JP4170247B2 - Image processing apparatus, image forming apparatus including the same, image processing method, program, and recording medium recording the program - Google Patents

Description

  The present invention relates to an image processing method and an image processing apparatus that determine a background from input image data and perform an optimum density correction process based on the determination result.

  Conventionally, when an original image is read by an image reading apparatus such as a scanner and the read image data is printed out by the image forming apparatus, if the read image data is output as it is by the image forming apparatus, background fog or show-through may occur. May occur. For this reason, it is a common practice to perform background removal processing to prevent background fog or show-through on the read image data before output by the image forming apparatus.

  For example, in Patent Document 1, a histogram of an input signal is created for each color component, and it is determined for each color whether or not at least two peaks of frequencies in the created histogram satisfy a predetermined condition. An image processing apparatus that performs background removal processing based on the result has been proposed.

Patent Document 2 proposes an image processing method for detecting an accurate background level in order to perform highly accurate background removal processing. More specifically, a histogram based on the luminance information of the input signal is created, a search is started from the dark side of the histogram, and a level first exceeding a predetermined threshold is detected as a background candidate level. An image processing method has been proposed in which the frequency is compared with the frequency of the level adjacent to the bright side, and the level with the higher frequency is determined as the background level.
JP 2000-22971 (Publication date: January 21, 2000) JP2003-51946 (release date: February 21, 2003)

  The background removal process is aimed at improving the legibility of a document with background fog or show-through and obtaining an optimal reproduction image, but it should be reproduced even if it is detected as the background in an actual document. There is a background that is not good and a base that is to be reproduced.

  For example, when the read original is an image such as a newspaper, the background is not a color that is intentionally added, but is caused by poor paper quality and yellowing due to amber color. Furthermore, since the paper quality is thin, show-through is likely to occur, and a background with little color, ie, near gray, is not preferable for legibility. For such a document, the background should be removed.

  On the other hand, in the case of a paper that is intentionally colored, such as a color paper, it should be reproduced as it is without removing the background. In the method of performing background detection without distinction, there is a problem that the background to be reproduced is not reproduced.

  In the configuration of Patent Document 1, the background of each color is determined. However, the background is removed when the background is detected regardless of the background to be reproduced and the background that should not be reproduced.

  The configuration of Patent Document 2 detects an accurate background level based on luminance information, but does not consider “color”.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to provide an image processing apparatus, an image forming apparatus, and an image processing apparatus capable of determining a highly accurate background in consideration of “color”. An image forming apparatus, an image processing method, a program, and a recording medium on which the program is recorded are provided.

  In order to solve the above problems, an image processing apparatus according to the present invention provides each pixel with at least an image processing apparatus that determines the presence or absence of the background of the document from the image data of each pixel obtained by reading the document. A maximum value calculating means for calculating for each pixel a maximum value calculated based on a calculation result using a pixel value of each color component, and a maximum value histogram creating means for creating a histogram relating to the maximum value based on the maximum value. And background determination means for determining the presence / absence of a background based on the histogram relating to the maximum value.

  By the way, the ground includes a ground that should be reproduced and a ground that should be removed without being reproduced. That is, in the case of paper with “color” that is intentionally colored like color paper, the background should be reproduced, while the background with poor paper quality such as newspaper, fading, yellowing, The gray background that hinders readability should be removed.

  Here, image data obtained by reading a document will be described as CMY data obtained by complementing RGB data with complementary colors. First, using the C, M, and Y values that are the pixel values of each color component, the absolute value of the difference between the color components, | C−M |, | M−Y |, and | Y−C | are calculated. The fact that all of the absolute values of the color component differences are small indicates an achromatic color in which the values of C, M, and Y are substantially the same. On the other hand, any one of the absolute values of the color component differences being a large value indicates a chromatic color. That is, the magnitude of “color” is determined by the magnitude of the maximum absolute value of the difference between the color components, which is a calculation result using the pixel values of the respective color components when attention is paid to a certain pixel. More specifically, “color tone” increases as the maximum value increases, and “color” decreases as the maximum value decreases. Therefore, considering the entire image, the background should be reproduced when the number of pixels having a large maximum value is large, and the background should be removed when the number of pixels having a small maximum value is large.

  Therefore, after creating a histogram related to the maximum value using the maximum value histogram creation means, the background judgment means determines whether or not there is a background based on the maximum value category and the frequency relationship in each category in the maximum value histogram. To do.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  In addition, by performing output based on the determination result of the presence or absence of the background, an output result with high legibility can be obtained in consideration of “color”.

  In order to solve the above problems, an image processing apparatus according to the present invention provides each pixel with at least an image processing apparatus that determines the presence or absence of the background of the document from the image data of each pixel obtained by reading the document. A maximum value calculating means for calculating a maximum value for each pixel among the pixel values of each color component, and a minimum value calculating means for calculating a minimum value for each pixel among the pixel values of each color component given to each pixel; Based on difference calculation means for calculating the difference between the maximum value and the minimum value in the same pixel for each pixel, difference histogram creation means for creating a histogram relating to the difference based on the difference, and a histogram relating to the difference And background determination means for determining the presence or absence of the background.

  By the way, the ground includes a ground that should be reproduced and a ground that should be removed without being reproduced. That is, in the case of paper with “color” that is intentionally colored like color paper, the background should be reproduced, while the background with poor paper quality such as newspaper, fading, yellowing, The gray background that hinders readability should be removed.

  Here, the magnitude of “color” is determined by the difference between the maximum value and the minimum value when attention is paid to a certain pixel. More specifically, the larger the difference, the larger the “color”, and the smaller the difference, the “color” becomes smaller. Therefore, considering the entire image, the background should be reproduced when the number of pixels with a large difference is large, and the background should be removed when the number of pixels with a small difference is large.

  Therefore, first, a difference between the maximum value calculated by the maximum value calculating means and the minimum value calculated by the minimum value calculating means is calculated by the difference calculating means. Then, after creating a histogram relating to the difference by the difference histogram creating means, the background determining means determines the presence / absence of the background based on the difference category and the frequency relationship in each section in the difference histogram.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  In addition, by performing output based on the determination result of the presence or absence of the background, an output result with high legibility can be obtained in consideration of “color”.

  In order to solve the above problems, an image processing apparatus according to the present invention provides each pixel with at least an image processing apparatus that determines the presence or absence of the background of the document from the image data of each pixel obtained by reading the document. A maximum value calculating means for calculating for each pixel a maximum value calculated based on a calculation result using a pixel value of each color component, and a calculation result using a pixel value of each color component given to each pixel A minimum value calculating means for calculating a minimum value for each pixel, a difference calculating means for calculating a difference between the maximum value and the minimum value in the same pixel for each pixel, and a histogram relating to the difference based on the difference A difference histogram creating means for creating a background, and a ground determination means for determining the presence or absence of a ground based on the histogram relating to the difference.

  By the way, the ground includes a ground that should be reproduced and a ground that should be removed without being reproduced. That is, in the case of paper with “color” that is intentionally colored like color paper, the background should be reproduced, while the background with poor paper quality such as newspaper, fading, yellowing, The gray background that hinders readability should be removed.

  Here, image data obtained by reading a document will be described as CMY data obtained by complementing RGB data with complementary colors. First, using the C, M, and Y values that are the pixel values of each color component, the absolute value of the difference between the color components, | C−M |, | M−Y |, and | Y−C | are calculated. The fact that all of the absolute values of the color component differences are small indicates an achromatic color in which the values of C, M, and Y are substantially the same. On the other hand, any one of the absolute values of the color component differences being a large value indicates a chromatic color. As a method of determining the magnitude of “color”, when paying attention to a certain pixel, not the maximum absolute value of the color component difference described above, but the maximum and minimum absolute values of the color component difference You may make it determine by the magnitude of a difference. More specifically, the larger the difference between the maximum and minimum absolute values of the color component difference calculated based on the calculation result using the pixel values of each color component, the greater the “color tone” and the smaller the difference. “Color tone” becomes smaller. Therefore, considering the entire image, the background should be reproduced when the number of pixels having a large difference is large, and the background should be removed when the number of pixels having a small difference is large.

  Therefore, first, the difference calculation unit calculates the difference between the maximum absolute value of the color component difference calculated by the maximum value calculation unit and the minimum absolute value of the color component difference calculated by the minimum value calculation unit. . Then, after creating a histogram relating to the difference by the difference histogram creating means, the background determining means determines the presence / absence of the background based on the difference category and the frequency relationship in each section in the difference histogram.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  In addition, by performing output based on the determination result of the presence or absence of the background, an output result with high legibility can be obtained in consideration of “color”.

  Further, the image processing apparatus according to the present invention is a color component histogram creating means for creating a histogram relating to the pixel value of each color component based on the pixel value of each color component of each pixel in the above image processing apparatus. The background determination means calculates a background level based on a histogram regarding the maximum value and a histogram regarding the pixel value of each color component.

  According to the above configuration, the background level is calculated by the background determination unit based on the histogram regarding the maximum value and the histogram regarding the pixel value of each color component.

  Therefore, it is possible to calculate the background level with higher accuracy than when the background level is calculated using only the histogram relating to the maximum value.

  In calculating the background level, for example, the background color may be determined using a histogram relating to the maximum value, and density information obtained from the histogram relating to the pixel value of each color component may be referred to.

  Further, the image processing apparatus according to the present invention is the above-described image processing apparatus, wherein the brightness value calculating means for calculating the brightness value of each pixel and the brightness for creating a histogram relating to the brightness value based on the brightness value of each pixel are provided. Histogram generating means, wherein the background determining means calculates a background level based on the histogram relating to the maximum value and the histogram relating to the luminance value.

  According to the above configuration, the background level is calculated by the background determination unit based on the histogram regarding the maximum value and the histogram regarding the luminance value.

  Therefore, it is possible to calculate the background level with higher accuracy than when the background level is calculated using only the histogram relating to the maximum value.

  In calculating the background level, for example, the color of the background may be determined using a histogram relating to the maximum value, and the luminance information obtained from the histogram relating to the luminance value may be referred to.

  Further, the image processing apparatus according to the present invention is a color component histogram creating means for creating a histogram relating to the pixel value of each color component based on the pixel value of each color component of each pixel in the above image processing apparatus. The background determination means calculates a background level based on a histogram regarding the difference and a histogram regarding the pixel value of each color component.

  According to the above configuration, the background level is calculated by the background determination unit based on the difference histogram and the histogram relating to the pixel value of each color component.

  Therefore, it is possible to calculate the background level with higher accuracy than when the background level is calculated using only the difference histogram.

  In calculating the background level, for example, the background color may be determined using a difference histogram, and further, density information obtained from the histogram relating to the pixel value of each color component may be referred to.

  Further, the image processing apparatus according to the present invention is the above-described image processing apparatus, wherein the brightness value calculating means for calculating the brightness value of each pixel and the brightness for creating a histogram relating to the brightness value based on the brightness value of each pixel are provided. Histogram generation means, wherein the background determination means calculates a background level based on a histogram relating to the difference and a histogram relating to the luminance value.

  According to the above configuration, the background level is calculated by the background determination unit based on the difference histogram and the histogram related to the luminance value.

  Therefore, it is possible to calculate the background level with higher accuracy than when the background level is calculated using only the difference histogram.

  In calculating the background level, for example, the background color may be determined using a difference histogram, and the luminance information obtained from the histogram related to the luminance value may be referred to.

  In the image processing apparatus according to the present invention, in the above image processing apparatus, the background determination unit may determine a background color from a predetermined feature amount based on a histogram regarding the maximum value and a histogram regarding the pixel value of each color component. Information on color or color and density is extracted, and background level correction is performed to correct the background level based on information on color or color and density extracted by the background determination means. It is characterized by having means.

  According to the above configuration, the background determination means extracts information about the color of the background, or the color and density from the predetermined feature amount. Here, the predetermined feature amount is, for example, a detection section of a histogram related to the maximum absolute value of the difference between the color components, a detection section of the histogram related to a pixel value of each color component, and a detection in the histogram of each color component. It is a value calculated using a category. As a specific example, there is a minimum value that is a minimum value of detection sections in a histogram of each color component, or a minimum difference that is a minimum value of differences between detection sections extracted from the histogram of each color component. Then, by the background level correction means, the detection division of the histogram relating to the maximum value which is the extracted color, the minimum difference of the detection division values in the histogram of each color component, or the magnitude relationship between these detection divisions, or these The background level is corrected based on the minimum value of each detection section in the histogram of each color component, which is density information in addition to color.

  Therefore, the background level can be corrected according to the color of the background, or the color and density. Therefore, it is possible to perform the background removal process with an appropriate background removal amount.

  In the image processing apparatus according to the present invention, in the above image processing apparatus, the background determination unit may determine a background color from a predetermined feature amount based on a histogram regarding the difference and a histogram regarding the pixel value of each color component. Or background level correction means for correcting the background level based on the color or information about color and density extracted by the background determination means. It is characterized by having.

  According to the above configuration, the background determination means extracts information about the color of the background, or the color and density from the predetermined feature amount. Here, the predetermined feature amount is, for example, a detection section of a histogram relating to a difference, a value of a detection section in a histogram relating to a pixel value of each color component, a value of each detection section in each color component histogram, and a histogram of each color component. This is a value calculated using the detection category. As a specific example, there is a minimum value that is a minimum value of each detection section of the histogram of each color component, or a minimum difference that is a minimum value of a difference of each detection section extracted from the histogram of each color component. Then, by the background level correction means, the detection division of the histogram relating to the difference that is the extracted color, the minimum difference of the detection division of each histogram of each color component, or the magnitude relationship of each of these detection divisions, or these colors In addition, the background level is corrected based on the minimum value of each detection section of the histogram of each color component, which is density information.

  Therefore, the background level can be corrected according to the color of the background, or the color and density. Therefore, it is possible to perform the background removal process with an appropriate background removal amount. The difference is a difference between the maximum value and the minimum value of the absolute value of the difference between the pixel values of the color components, even if it is the difference between the maximum value and the minimum value of the pixel values of each color component. Also good.

  Further, in the image processing apparatus according to the present invention, in the above image processing apparatus, the background determination unit may determine a background color from a predetermined feature amount based on a histogram regarding the maximum value and a histogram regarding the luminance value. Alternatively, information relating to color and brightness is extracted, and background level correction means for correcting the background level based on the color extracted by the background determination means or information relating to color and brightness is provided. It is characterized by having.

  According to the configuration described above, the background determination unit extracts the background color or information about the color and luminance from the predetermined feature amount. Here, the predetermined feature amount is, for example, a histogram detection section related to the maximum absolute value of the color component difference or a histogram detection section related to the luminance of each pixel value. Then, by the background level correction means, based on the value of the detection division in the histogram relating to the maximum value which is the extracted color, or the detection division in the histogram relating to the luminance of each pixel value which is luminance information in addition to the color, The background level is corrected.

  Therefore, the background level can be corrected according to the color of the background, or the color and brightness. Therefore, it is possible to perform the background removal process with an appropriate background removal amount.

  In the image processing apparatus according to the present invention, in the above image processing apparatus, the background determination unit may calculate a background color from a predetermined feature amount based on a histogram related to the difference and a histogram related to the luminance value, or , Which extracts information relating to color and brightness, and comprises ground level correction means for correcting the background level based on the color extracted by the background determination means or information relating to color and brightness. It is characterized by being.

  According to the configuration described above, the background determination unit extracts the background color or information about the color and luminance from the predetermined feature amount. Here, the predetermined feature amount is, for example, a histogram detection section regarding a difference or a histogram detection section regarding the luminance of each pixel value. Then, based on the value of the detection section of the histogram related to the difference that is the extracted color by the background level correction means, or the value of the detection classification of the histogram related to the brightness of each pixel value that is luminance information in addition to the color. Level is corrected.

  Therefore, the background level can be corrected according to the color of the background, or the color and brightness. Therefore, it is possible to perform the background removal process with an appropriate background removal amount.

  According to the image processing apparatus of the present invention, in the above image processing apparatus, a shape determining unit that determines a shape of a histogram related to the maximum value and a shape of a histogram related to a pixel value of each color component, and the shape determining unit And a background level correcting means for correcting the background level according to the determination result.

  Here, when the show-through occurs in the document, the difference value of the area where the show-through occurs tends to be substantially the same value as the difference value of the area where the show-through does not occur. On the other hand, the pixel value of each color component in the area where the show-through occurs does not become the same value as the pixel value of each color component in the area where the show-through does not occur, and the pixel value of each color component tends to increase.

  Therefore, when show-through occurs, the shape of the histogram relating to the maximum value has a sharp shape, whereas the shape of the histogram relating to the pixel value of each color component tends to be a broad shape.

  Therefore, the presence / absence of show-through can be determined from the shape of the histogram related to the maximum value and the shape of the histogram related to the pixel value of each color component by the shape determination means.

  Therefore, by correcting the background level according to the presence or absence of show-through by the background level correction means, it is possible to perform background removal processing with an appropriate removal amount.

  An image processing apparatus according to the present invention includes: a shape determination unit that determines a shape of a histogram related to the difference and a shape of a histogram related to a pixel value of each color component; According to the determination result, a background level correcting unit for correcting the background level is provided.

  Here, when the show-through occurs in the document, the difference value of the area where the show-through occurs tends to be substantially the same value as the difference value of the area where the show-through does not occur. On the other hand, the pixel value of each color component in the area where the show-through occurs does not become the same value as the pixel value of each color component in the area where the show-through does not occur, and the pixel value of each color component tends to increase.

  Therefore, when show-through occurs, a histogram relating to the difference (maximum value and minimum value of the absolute value of the difference between the maximum and minimum values of the pixel values of each color component or the difference between the pixel values of the color components) On the other hand, the shape of the histogram related to the pixel value of each color component tends to be a broad shape.

  Therefore, the presence / absence of show-through can be determined from the shape of the histogram relating to the difference and the shape of the histogram relating to the pixel value of each color component by the shape determining means.

  Therefore, by correcting the background level according to the presence or absence of show-through by the background level correction means, it is possible to perform background removal processing with an appropriate removal amount.

  In the image processing apparatus according to the present invention, in the above-described image processing apparatus, a shape determination unit that determines a shape of a histogram related to the maximum value and a shape of the histogram related to the luminance value, and a determination result of the shape determination unit And a background level correcting means for correcting the background level.

  Here, when the show-through occurs in the document, the difference value of the area where the show-through occurs tends to be substantially the same value as the difference value of the area where the show-through does not occur. On the other hand, the brightness value of the area where the show-through has occurred is not the same as the brightness value of the area where the show-through has not occurred, and the brightness value tends to be low.

  Therefore, when show-through has occurred, the shape of the histogram relating to the maximum value has a sharp shape, while the shape of the histogram relating to the luminance value tends to be a broad shape.

  Therefore, the presence / absence of show-through can be determined from the shape of the histogram related to the maximum value and the shape of the histogram related to the luminance value by the shape determination means.

  Therefore, by correcting the background level according to the presence or absence of show-through by the background level correction means, it is possible to perform background removal processing with an appropriate removal amount.

  According to the image processing apparatus of the present invention, in the image processing apparatus described above, the shape determination unit that determines the shape of the histogram related to the difference and the shape of the histogram related to the luminance value, and the determination result of the shape determination unit Accordingly, a background level correcting means for correcting the background level is provided.

  Here, when the show-through occurs in the document, the difference value of the area where the show-through occurs tends to be substantially the same value as the difference value of the area where the show-through does not occur. On the other hand, the brightness value of the area where the show-through has occurred is not the same as the brightness value of the area where the show-through has not occurred, and the brightness value tends to be low.

  Therefore, if show-through has occurred, a difference histogram (the maximum and minimum values of the pixel values of each color component or the absolute value of the difference between the pixel values of the color components and the minimum value) The histogram of the luminance value has a tendency to be a broad shape.

  Therefore, the presence / absence of show-through can be determined by the shape determination means from the shape of the histogram related to the difference and the shape of the histogram related to the luminance value.

  Therefore, by correcting the background level according to the presence or absence of show-through by the background level correction means, it is possible to perform background removal processing with an appropriate removal amount.

  In the image processing apparatus according to the present invention, in the image processing apparatus described above, an area separation unit that separates the read original into a plurality of areas including a character area and / or a halftone dot area; A pixel selection unit that selects pixels in a predetermined region other than the character region and / or the dot region, and determines whether or not the background is present for the pixel selected by the pixel selection unit. It is characterized by that.

  Therefore, whether or not there is a background to be removed for pixels in a predetermined area other than the character area and / or the dot area, specifically, a pixel separated into a photographic area including a pixel forming the background. A determination is made. That is, pixels other than the above-described photographic area can be excluded from the determination target area in advance. Originals include only text, only printed photos (halftone area), mixed text and printed photos, mixed text / printed photos / photos (continuous tone areas such as silver halide photos), etc. . Pixels in a predetermined area other than the character area and / or halftone dot area are pixels that form a background, or pixels that form a background and pixels in a photographic area when these types of originals are targeted. Means.

  Accordingly, it is possible to delete data other than the area including the pixels forming the background, and the effect of improving the detection accuracy of the background and the processing speed of the image processing is achieved.

  The image processing apparatus according to the present invention is characterized in that, in the image processing apparatus described above, the background level is calculated for the pixel selected by the pixel selection means.

  Therefore, the background level is calculated for pixels in a predetermined region other than the character region and / or the dot region. That is, the character and halftone dot area not including the background to be removed can be excluded in advance from the target area for the background level calculation.

  Therefore, the base level can be calculated efficiently and in a short time.

  In order to solve the above problems, an image forming apparatus according to the present invention includes the above-described image processing apparatus.

  Therefore, the presence / absence of the background can be accurately determined and the background removal process can be performed, so that an effect that a high-quality image can be output can be obtained.

  In order to solve the above problems, an image processing method according to the present invention is provided to each pixel in an image processing method that determines at least the presence or absence of the background of the document from the image data of each pixel obtained by reading the document. A maximum value calculating step for calculating for each pixel a maximum value calculated based on a calculation result using pixel values of each color component; and a maximum value histogram creating step for creating a histogram relating to the maximum value based on the maximum value; And a background determination step for determining the presence or absence of a background based on a histogram relating to the maximum value.

  By the way, the ground includes a ground that should be reproduced and a ground that should be removed without being reproduced. That is, in the case of paper with “color” that is intentionally colored like color paper, the background should be reproduced, while the background with poor paper quality such as newspaper, fading, yellowing, The gray background that hinders readability should be removed.

  Here, image data obtained by reading a document will be described as CMY data obtained by complementing RGB data with complementary colors. First, using the C, M, and Y values that are the pixel values of each color component, the absolute value of the difference between the color components, | C−M |, | M−Y |, and | Y−C | are calculated. The fact that all of the absolute values of the color component differences are small indicates an achromatic color in which the values of C, M, and Y are substantially the same. On the other hand, any one of the absolute values of the color component differences being a large value indicates a chromatic color. That is, the magnitude of “color” is determined by the magnitude of the maximum absolute value of the difference between the color components, which is a calculation result using the pixel values of the respective color components when attention is paid to a certain pixel.

  Therefore, first, the absolute value of the difference between the color components, | C−M |, | MY− |, and | Y−C | are calculated from the pixel values of the respective color components given to the respective pixels. Next, the maximum absolute value of the color component difference is calculated for each pixel. Then, after creating a histogram relating to the maximum absolute value of the difference between the color components, the presence or absence of a background is determined based on the frequency relationship in each section of the histogram.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  In addition, by performing output based on the determination result of the presence or absence of the background, an output result with high legibility can be obtained in consideration of “color”.

  Also, the image processing method according to the present invention provides at least the pixel value of each color component given to each pixel in the image processing method for determining the presence or absence of the background of the document from the image data of each pixel obtained by reading the document. Among these, a maximum value calculating step for calculating a maximum value for each pixel, a minimum value calculating step for calculating a minimum value for each pixel among pixel values of each color component given to each pixel, and the maximum value in the same pixel A difference calculating step for calculating a difference from the minimum value for each pixel, a difference histogram creating step for creating a histogram relating to the difference based on the difference, and the presence / absence of a background based on the histogram relating to the difference And a background determination step.

  By the way, the ground includes a ground that should be reproduced and a ground that should be removed without being reproduced. That is, in the case of paper with “color” that is intentionally colored like color paper, the background should be reproduced, while the background with poor paper quality such as newspaper, fading, yellowing, The gray background that hinders readability should be removed.

  Here, the magnitude of “color” is determined by the difference between the maximum value and the minimum value when attention is paid to a certain pixel. More specifically, the larger the difference, the larger the “color”, and the smaller the difference, the “color” becomes smaller. Therefore, considering the entire image, the background should be reproduced when the number of pixels with a large difference is large, and the background should be removed when the number of pixels with a small difference is large.

  Here, first, among the pixel values of each color component, the maximum value and the minimum value are calculated for each pixel, and the difference between the maximum value and the minimum value in the same pixel is obtained. Then, a histogram relating to the difference is created. Next, based on the histogram relating to the difference, the presence / absence of a background is determined based on the relationship between the category (detection category) and the frequency in each category (detection category).

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  In addition, by performing output based on the determination result of the presence or absence of the background, an output result with high legibility can be obtained in consideration of “color”.

  Also, the image processing method according to the present invention provides at least the pixel value of each color component given to each pixel in the image processing method for determining the presence or absence of the background of the document from the image data of each pixel obtained by reading the document. A maximum value calculating step for calculating for each pixel a maximum value calculated based on the calculation result used, and a minimum value calculated based on the calculation result using the pixel value of each color component given to each pixel A minimum value calculating step for calculating the difference, a difference calculating step for calculating the difference between the maximum value and the minimum value in the same pixel for each pixel, and a difference histogram creating step for creating a histogram relating to the difference based on the difference And background determination for determining the presence / absence of a background based on the histogram relating to the difference.

  By the way, the ground includes a ground that should be reproduced and a ground that should be removed without being reproduced. That is, in the case of paper with “color” that is intentionally colored like color paper, the background should be reproduced, while the background with poor paper quality such as newspaper, fading, yellowing, The gray background that hinders readability should be removed.

  Here, image data obtained by reading a document will be described as CMY data obtained by complementing RGB data with complementary colors. First, using the C, M, and Y values that are the pixel values of each color component, the absolute value of the difference between the color components, | C−M |, | M−Y |, and | Y−C | are calculated. The fact that all of the absolute values of the color component differences are small indicates an achromatic color in which the values of C, M, and Y are substantially the same. On the other hand, any one of the absolute values of the color component differences being a large value indicates a chromatic color. As a method of determining the magnitude of “color”, when paying attention to a certain pixel, not the maximum absolute value of the color component difference described above, but the maximum and minimum absolute values of the color component difference You may make it determine by the magnitude of a difference. More specifically, the larger the difference between the maximum and minimum absolute values of the color component difference calculated based on the calculation result using the pixel values of each color component, the greater the “color tone” and the smaller the difference. “Color tone” becomes smaller. Therefore, considering the entire image, the background should be reproduced when the number of pixels having a large difference is large, and the background should be removed when the number of pixels having a small difference is large.

  Therefore, first, the maximum and minimum absolute values of the color component difference are calculated, and the difference between the maximum and minimum values is obtained. Then, a histogram relating to the difference is created, and the presence / absence of a background is determined based on the difference category and the frequency relationship in each category.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  In addition, by performing output based on the determination result of the presence or absence of the background, an output result with high legibility can be obtained in consideration of “color”.

  As described above, the program according to the present invention is for causing a computer to function as each unit of the image processing apparatus.

  By loading the program into the computer system, the image processing apparatus can be provided to the user.

  The recording medium recording the program according to the present invention records the program as described above.

  By loading the program recorded on the recording medium into a computer system, the image processing apparatus or the image processing method can be provided to the user.

  As described above, the image processing apparatus according to the present invention includes, as described above, a maximum value calculation unit that calculates, for each pixel, a maximum value calculated based on a calculation result using a pixel value of each color component given to each pixel, A maximum value histogram creating means for creating a histogram relating to the maximum value based on the maximum value, and a background determining means for judging the presence or absence of the background based on the histogram relating to the maximum value.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  Further, as described above, the image processing apparatus according to the present invention includes a maximum value calculation unit that calculates a maximum value for each pixel among the pixel values of each color component given to each pixel, and each color component given to each pixel. Based on the difference, a minimum value calculating means for calculating a minimum value for each pixel, a difference calculating means for calculating a difference between the maximum value and the minimum value in the same pixel for each pixel, and A difference histogram creating means for creating a histogram relating to the difference, and a background determining means for determining the presence or absence of the background based on the histogram relating to the difference.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  Further, as described above, the image processing apparatus according to the present invention includes a maximum value calculating unit that calculates, for each pixel, a maximum value calculated based on a calculation result using a pixel value of each color component given to each pixel. A minimum value calculating means for calculating for each pixel a minimum value calculated based on a calculation result using a pixel value of each color component given to each pixel; and a difference between the maximum value and the minimum value in the same pixel. A difference calculating means for calculating for each pixel; a difference histogram creating means for creating a histogram related to the difference based on the difference; and a ground determining means for determining the presence or absence of the ground based on the histogram related to the difference. Yes.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  As described above, the image forming apparatus according to the present invention includes the above-described image processing apparatus. Therefore, it is possible to determine the presence / absence of a highly accurate background in consideration of “color”, and to obtain an output result with high legibility considering “color”.

  In the image processing method according to the present invention, as described above, the maximum value calculation step for calculating, for each pixel, the maximum value calculated based on the calculation result using the pixel value of each color component given to each pixel; A maximum value histogram creating step for creating a histogram relating to the maximum value based on the maximum value, and a background determining step for judging the presence or absence of the background based on the histogram relating to the maximum value.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  The image processing method of the present invention includes a maximum value calculating step for calculating a maximum value for each pixel among the pixel values of each color component given to each pixel, and among the pixel values of each color component given to each pixel, A minimum value calculating step for calculating a minimum value for each pixel, a difference calculating step for calculating a difference between the maximum value and the minimum value in the same pixel for each pixel, and a histogram relating to the difference based on the difference A difference histogram creating step, and a background determination step for determining the presence or absence of a background based on the histogram relating to the difference.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  Further, the image processing method of the present invention is provided to each pixel, and a maximum value calculating step for calculating, for each pixel, a maximum value calculated based on a calculation result using a pixel value of each color component given to each pixel. A minimum value calculation step for calculating, for each pixel, a minimum value calculated based on a calculation result using pixel values of each color component, and a difference between the maximum value and the minimum value in the same pixel is calculated for each pixel. It is characterized by comprising a difference calculating step, a difference histogram creating step for creating a histogram related to the difference based on the difference, and a background determining step for determining the presence or absence of the background based on the histogram related to the difference.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  As described above, the program according to the present invention is for causing a computer to function as each unit of the image processing apparatus.

  Therefore, it is possible to provide the user with the image processing apparatus by loading the program into the computer system.

  The recording medium recording the program according to the present invention records the program as described above.

  Therefore, by loading the program recorded on the recording medium into the computer system, the image processing apparatus or the image processing method can be provided to the user.

[Embodiment 1]
An embodiment of the present invention will be described below with reference to FIGS.

  FIG. 2 is a cross-sectional view of the main part showing a schematic configuration of the color image forming apparatus. 2. Description of the Related Art In recent years, in color image forming apparatuses, a tandem type apparatus as shown in FIG. 2 is becoming mainstream in combination with downsizing and unitization of internal apparatuses to make them relatively inexpensive.

  The color image forming apparatus includes image forming stations 100K, 100C, 100M, and 100Y corresponding to K (black), C (cyan), M (magenta), and Y (yellow). For example, in the K image forming station 100K, the drum-shaped photoconductor 101K rotates in the direction of the arrow D in the figure, and there are a charger 102K, a writing optical system 103K, a developing device 104K, and a cleaning device 105K at least in the order of rotation. Is arranged.

  In the color image forming apparatus, the surface of the photosensitive member between the charger 102K and the developing device 104K is irradiated with laser light from the writing optical system 103K, and an electrostatic latent image corresponding to K image data is formed on the photosensitive member 101K. It has come to be.

  The image forming stations 100C, 100M, and 100Y have the same configuration as that of the image forming station 100K. The chargers 102C, 102M, and 102Y, the writing optical systems 103C, 103M, and 103Y, and the developing device are centered on the photosensitive members 101C, 101M, and 101Y. 104C / 104M / 104Y and cleaning devices 105C / 105M / 105Y are arranged. That is, each of the image forming stations K, C, M, and Y has the same configuration except that the toner color inside the developing device is different.

  The image forming stations 100K, 100C, 100M, and 100Y are juxtaposed along the paper conveying belt 110, and the paper conveying belt 110 is disposed between the developing device and the cleaning device in each of the K, C, M, and Y image forming stations. In contact with the photosensitive member.

  Transfer rollers 106K, 106C, 106M, and 106Y for applying a transfer bias in each of the image forming stations 100K, 100C, 100M, and 100Y are disposed on the surface (rear surface) that contacts the back side of each photoconductor side of the sheet conveying belt 110. Yes.

  In the color image forming apparatus having the configuration shown in FIG. 2, the image forming operation is performed as follows. First, in each image forming station, the photosensitive member is charged by a charger, and then an electrostatic latent image corresponding to each color image to be created is formed by laser light in a writing optical system.

  Next, the electrostatic latent image is developed by a developing device to form a toner image. The developing devices are developing devices that perform development with toners of K, C, M, and Y, respectively, and the toner images of the respective colors formed on the four photoconductors are superimposed on the paper.

  The paper is fed onto the paper transport belt 110 from the paper feed unit 120 in the direction of arrow E in the drawing in synchronism with the image formation on the photoconductor. The paper held on the paper transport belt 110 is transported, and each color toner image is transferred at a contact position (transfer portion) with each photoconductor. The toner image on the photoconductor is transferred onto the sheet by an electric field formed by the potential difference between the transfer bias applied to the transfer roller and the photoconductor.

  The recording paper on which the four color toner images are passed through the four transfer portions is conveyed to the fixing device 130, where the toner is fixed, and is discharged to a paper discharge portion (not shown). Further, residual toner that is not transferred by the transfer unit and remains on each photoconductor is collected by a cleaning device.

  In the example of FIG. 2, the image forming stations are arranged in the order of K, C, M, and Y from the upstream side to the downstream side in the paper conveyance direction. However, the order is not limited to this order. Is arbitrarily set. In addition, as described above, each image forming station has the same configuration except that the toner color in the developing device is different, and the rest of the configuration is the same. Is set.

  The color image forming apparatus can be combined with a color scanner device to constitute a color copying machine. A configuration of a digital color copying machine as an embodiment including a color image processing apparatus to which a color image processing method according to the present invention is applied will be described below.

  As shown in FIG. 3, the digital color copying machine includes a color image processing apparatus (image processing apparatus) 1 and a color image input apparatus 2 and a color image output apparatus 3 connected to form a digital color copying machine as a whole. To do. Further, the color image processing apparatus 1 can input various settings from the operation panel 4 for setting the operation of the digital color copying machine. The operation panel 4 includes a display unit such as a liquid crystal display, buttons for setting a copy button, a numeric keypad, an image mode such as a character / photo mode, and a copy mode such as a color copy / monochrome copy.

  The color image input device 2 is composed of, for example, a scanner unit having a CCD (Charge Coupled Device), and the reflected light image from the original is converted into an CCD (R: red, G: green, B: blue) analog signal as a CCD. And input to the color image processing apparatus.

  The color image processing apparatus 1 includes an A / D conversion unit 10, a shading correction unit 11, a region separation processing unit 12, an input tone correction unit 13, a color correction unit 14, a black generation and under color removal unit 15, and a spatial filter processing unit 16. , An output tone correction unit 17 and a tone reproduction processing unit 18. The analog signal read by the color image input device 2 is converted into an A / D conversion unit 10, a shading correction unit 11, a region separation processing unit (region separation means) 12, an input gradation in the color image processing device 1. The correction unit 13, the color correction unit 14, the black generation and under color removal unit 15, the spatial filter processing unit 16, the output gradation correction unit 17, and the gradation reproduction processing unit 18 are sent in this order and converted into CMYK digital color signals. And output to the color image output device 3.

  The A / D converter 10 converts RGB analog signals read by the color image input device 2 into RGB digital signals. Further, the shading correction unit 11 removes various distortions generated in the illumination system, the imaging system, and the imaging system of the color image input device 2 from the RGB digital signal, and the RGB reflectance signal. Thus, the color balance is adjusted, and the density signal is converted into a signal that can be easily handled by the image processing system employed in the color image processing apparatus 1.

  Based on the RGB signal input from the shading correction unit 11, the region separation processing unit 12 converts each pixel in the input image into a third region (photo region) other than the character region, halftone dot region, character region, and halftone dot region. It isolate | separates into either. In the region separation processing unit 12, based on the separation result, a region identification signal indicating which region each pixel belongs to is input to the input gradation correction unit 13, the black generation and under color removal unit 15, the spatial filter processing unit 16, The RGB signal input from the shading correction unit 11 is output to the subsequent color correction unit 14 as it is. The area separation function is not limited to these, and each pixel in the input image is not limited to the character area, halftone dot area, and photographic area, but is also a background area, a blurred area, or a halftone dot character. You may make it isolate | separate into.

  The input gradation correction unit 13 performs input gradation correction such as background determination and background removal. The configuration and processing of the input tone correction unit 13 will be described in detail later.

  The color correction unit 14 performs processing for removing color turbidity based on the spectral characteristics of CMY (C: cyan, M: magenta, Y: yellow) color materials including unnecessary absorption components in order to realize faithful color reproduction. .

  The black generation and under color removal unit 15 generates black (K) signals from the CMY three-color signals after color correction, and subtracts the K signals obtained by black generation from the original CMY signals to generate new CMY signals. Generate the process. As a result, the CMY three-color signal is converted into a CMYK four-color signal in the black generation and lower color removal unit 15.

  As a general method of black generation processing, there is a method of generating black by skeleton black. In this method, the input / output characteristic of the skeleton curve is y = f (x), the input data is C, M, Y, the output data is C ′, M ′, Y ′, K ′, UCR (Under Color When the removal rate is α (0 <α <1), the black generation and under color removal processing is expressed by the following equation (1).

  The spatial filter processing unit 16 performs spatial filter processing using a digital filter on the image data of the CMYK signal input from the black generation and under color removal unit 15 based on the region identification signal, thereby correcting the spatial frequency characteristics. Processing is performed to prevent blurring of the output image and deterioration of graininess. Similar to the spatial filter processing unit 16, the gradation reproduction processing unit 18 performs predetermined processing on the image data of the CMYK signal based on the region identification signal.

  For example, the region separated into characters by the region separation processing unit 12 is a high-frequency enhancement amount by sharp enhancement processing in the spatial filter processing in the spatial filter processing unit 16 in order to improve the reproducibility of black characters or color characters in particular. Is increased. At the same time, the tone reproduction processing unit 18 selects binarization or multi-value processing on a high-resolution screen suitable for high-frequency reproduction.

  Further, with respect to the region separated into halftone dots by the region separation processing unit 12, the spatial filter processing unit 16 performs low-pass filter processing for removing the input halftone component. The output tone correction unit 17 performs an output tone correction process for converting a signal such as a density signal into a halftone dot area ratio that is a characteristic value of the color image output device. Gradation reproduction processing (halftone generation) is performed so that the image is finally separated into pixels and each gradation is reproduced. For the region separated into photographs by the region separation processing unit 12, binarization or multi-value processing is performed on the screen with an emphasis on gradation reproducibility.

  The image data that has been subjected to the above-described processes is temporarily stored in a storage unit, read at a predetermined timing, and input to the color image output device 3.

  The color image output device 3 outputs image data onto a recording medium (for example, paper). For example, an electrophotographic color image forming device as shown in FIG. 2 or a color image formation using an ink jet method is used. Although an apparatus etc. can be mentioned, it is not specifically limited. The above processing is controlled by a CPU (Central Processing Unit) (not shown).

  Next, FIG. 1 shows the configuration of the input tone correction unit 13 that performs image processing according to the features of the present invention.

  The input tone correction unit 13 includes a maximum value calculation unit (maximum value calculation unit) 30, a minimum value calculation unit (minimum value calculation unit) 31, a difference calculation unit (difference calculation unit) 32, and a difference histogram creation unit (difference histogram calculation). Means) 33, a determination unit (background determination unit) 34, a conversion table selection unit 35, and a density conversion unit 36.

  In the following description, RGB data as an input signal is CMY data (C = 255-R, M = 255-G, Y = 255-B) that has been subjected to complementary color inversion by a complementary color inversion unit (not shown). Later, it is assumed that the processing in each of the members 30 to 36 is performed. However, the present invention is not limited to this, and a configuration in which the processing of each member 30 to 36 is performed on RGB data that is an input signal may be employed. The complementary color inversion unit may be provided in the input tone correction unit 13 or may be provided between the region separation processing unit 12 and the input tone correction unit 13.

  Each value of C, M, Y, R, G, and B corresponds to the pixel value of each color component described in the means for solving the problem.

  Further, as a histogram relating to “color”, (a) the maximum value max {| CM |, | MY |, | YC |} of the absolute value of the difference between C, M, and Y, which is the pixel value of each color component, (B) The difference max (C, M, Y) -min (C, M, Y) between the maximum and minimum pixel values of each color component, and (c) C · M · A histogram based on the difference between the maximum and minimum absolute value of the difference of Y max {| CM |, | MY |, | YC |} -min {| CM |, | MY |, | YC |} However, in the following description, a histogram based on the difference between the maximum value and the minimum value of the pixel values of each color component in (b) will be described as an example.

  The maximum value calculation unit 30 calculates the maximum values of C, M, and Y for each pixel after the above complementary color inversion. The minimum value calculation unit 31 calculates the minimum values of C, M, and Y for each pixel after the complementary color inversion. Further, the difference calculation unit 32 calculates a difference between the maximum value calculated by the maximum value calculation unit 30 and the minimum value calculated by the minimum value calculation unit 31.

  For example, it is assumed that the values of C, M, and Y of a certain pixel are (C, M, Y) = (200, 150, 80). In this case, the maximum value calculation unit 30 calculates 200 which is the maximum value among C, M, and Y. The minimum value calculation unit calculates 80, which is the minimum value among C, M, and Y. Then, the difference calculation unit 32 calculates 120 which is a value obtained by subtracting 80 which is the minimum value from 200 which is the maximum value (hereinafter referred to as a difference value).

  The difference histogram creation unit 33 creates a histogram (histogram related to difference) based on the difference value. The determination unit 34 determines whether or not there is a ground based on the information of the histogram. A specific determination method will be described later.

  The conversion table selection unit 35 selects a conversion table according to the determination result of the determination unit 34. Specifically, when the determination unit 34 determines that there is a background, a conversion table for background removal (hereinafter referred to as a background removal table) is selected, and the determination unit 34 determines that there is no background. If so, a conversion table for no background removal (hereinafter referred to as a table without background removal) is selected.

  The density conversion unit 36 uses the conversion table selected by the conversion table selection unit 35 to perform a density conversion process on the pixel image data after the complementary color inversion. Then, the data whose density has been converted by the density converter 36 is sent to the color corrector 14. Note that the density conversion table is provided in the density conversion unit 36 itself. However, the present invention is not limited to this, and the above-described density conversion table may be provided in a storage unit (not shown) and called from the storage unit each time conversion is performed. The same applies to the following second and third embodiments. The density conversion specifically refers to conversion of pixel values of each color component.

  The present invention is not limited to the above-described configuration, and it may be configured to perform pre-scanning, determine the background from the information, and perform density correction processing in the main scanning. In this case, during the main scan, processing in the maximum value calculation unit 30, the minimum value calculation unit 31, the difference calculation unit 32, the difference histogram creation unit 33, the determination unit 34, and the conversion table selection unit 35 is not performed, and the density conversion unit 36 is performed. Only the processing at is performed.

  Also, the histogram may be created for all pixels, or may be created by thinning out pixels. In the case of a pre-scan configuration, a method of reducing the processing time and pixel data by changing the scan speed may be adopted.

  Next, a method for determining whether or not there is a ground in the determination unit 34 will be described. As this determination method, for example, a technique disclosed in Patent Document 2 (Japanese Patent Laid-Open No. 2000-51946) shown in the related art can be used. The method for determining whether or not there is a background in Patent Document 2 will be described as follows.

  A background determination process is performed by creating a histogram from the input image data and comparing the frequency of the histogram with a plurality of predetermined threshold values. At this time, as a first step, the histogram is searched in a predetermined direction, and the frequency of the density category is compared with the first threshold value to determine whether or not to perform the subsequent processing. As a second step, when it is determined that the subsequent processing is performed in the first step, it is determined whether or not there is a background candidate by comparing the frequency of the density division with the second threshold value. As a third step, when it is determined that there is a background candidate in the second step, the frequency of the density section whose frequency is equal to or higher than the second threshold and the direction adjacent to the density section and in the search direction The presence / absence of the background is determined by adding the frequency of the density category to be compared and comparing with the third threshold value.

  Here, in the present embodiment, “density category” in this determination method is considered as “difference value category”. Further, for convenience of explanation, “frequency” is considered as “ratio (%)”.

  By the way, in the case of the present embodiment, as described above, the difference histogram is created based on the difference value in the pixel for each pixel. Here, a large difference value means that “color” is large, and a small difference value means that “color” is small.

  In other words, in the case of a “colored” background that is intentionally colored like color paper, the ratio of pixels having a large difference value is high in the histogram. On the other hand, in the case of a background with poor paper quality, such as newspaper, or a gray background that interferes with fading, yellowing, and legibility, the ratio of pixels with small difference values increases.

  Therefore, in the present embodiment, it is determined whether or not there is a ground in consideration of the above characteristics. Hereinafter, a description will be given with reference to FIG.

  FIG. 4 is an example of a difference histogram created by the difference histogram creation unit 33. In this difference histogram, as shown in the figure, 1 to 32 are classified according to the difference value. Specifically, classification is performed such that if the difference value is large, the classification value also increases. The vertical axis indicates the ratio (%) of each segment to all pixels.

  Here, the detection direction on the difference histogram is determined from the smaller difference value to the larger difference value as shown by the arrow direction in FIG. That is, the detection direction is determined from the smaller value of the classification to the larger value. In addition, the upper limit value of the classification for detection is determined. Up to this upper limit value, it is determined that there is a background when a predetermined condition is satisfied, and it is determined that there is no background when the predetermined condition is not satisfied.

  For example, the upper limit is set to Category 5, and further using the determination method shown in Patent Document 2 described above, the first threshold is 3% of all pixels, the second threshold is 5% of all pixels, and The threshold value of 3 is 15% of all pixels.

  First, in Category 1, as shown in the figure, since the ratio (1.5%) of Category 1 is smaller than the first threshold (3%), it is determined that the subsequent processing is performed, and the second Proceed to the process. Advancing by one in the search direction (arrow direction in the figure), the ratio of the category 2 is compared with the second threshold. Since the ratio of Category 2 (1.5%) is smaller than the second threshold (5%), the process proceeds one more in the search direction (arrow direction in the figure) without performing the third step. A comparison is made between the proportion of category 3 and the second threshold.

  Since the ratio of category 3 (44.1%) is larger than the second threshold (5%), it is determined that there is a background candidate.

  In addition, since it is determined that there is a background candidate, a comparison between the total of the ratio of Category 3 (44.1%) and the ratio of Category 4 (14.7%) and the third threshold (15%), That is, the third step is performed. Here, since the above sum is larger than the third threshold value, in Category 3, the predetermined condition in this determination method is satisfied, and it is determined that there is a background.

  As described above, since it is determined that there is a background in a section smaller than the upper limit section 5, the density conversion unit 36 performs density conversion processing using the above background removal table. On the other hand, for example, when it is determined that there is no background in a section smaller than the upper limit section 5, the density conversion unit 36 performs density conversion processing using the above background removal-free table. . In the following, including the second and third embodiments to be described later, a category that satisfies the predetermined condition is referred to as a detection category relating to a difference.

  FIG. 5 is a flowchart showing an operation example of the input tone correction unit 13. In the present embodiment, the difference calculation unit 32 calculates the difference between the maximum value and the minimum value for each pixel based on the CMY data that has been subjected to complementary color inversion (S1). After S1, the difference histogram creation unit 33 creates a difference histogram based on the difference value (S2). After S2, the determination unit 34 determines whether or not there is a background from the created difference histogram (S3).

  If the determination unit 34 determines that there is a background in S3, the conversion table selection unit 35 selects a background removal table (S4). On the other hand, when the determination unit 34 determines that there is no background, the conversion table selection unit 35 selects a table without background removal (S5).

  After S4 or S5, the density conversion unit 36 performs density conversion processing using the selected conversion table (S6). Thus, a series of processing ends.

  As described above, the image processing apparatus according to the present embodiment includes a maximum value calculation unit (maximum value calculation unit) 30 that calculates a maximum value for each pixel among pixel values of each color component given to each pixel, Of the pixel values of each color component given to each pixel, a minimum value calculation unit (minimum value calculation means) 31 that calculates the minimum value for each pixel, and the difference between the maximum value and the minimum value in the same pixel A difference calculation unit (difference calculation unit) 32 that calculates the difference, a difference histogram generation unit (difference histogram generation unit) 33 that generates a histogram related to the difference based on the difference, and a background histogram based on the histogram related to the difference A determination unit (background determination unit) 34 that determines presence / absence is provided.

  By the way, the ground includes a ground that should be reproduced and a ground that should be removed without being reproduced. That is, in the case of paper with “color” that is intentionally colored like color paper, the background should be reproduced, while the background with poor paper quality such as newspaper, fading, yellowing, The gray background that hinders readability should be removed.

  Here, the magnitude of “color” is determined by the difference between the maximum value and the minimum value when attention is paid to a certain pixel. More specifically, the larger the difference, the larger the “color”, and the smaller the difference, the “color” becomes smaller. Therefore, considering the entire pixel, the background should be reproduced when the number of pixels with a large difference is large, and the background should be removed when the number of pixels with a small difference is large.

  Therefore, the difference histogram creation unit 33 creates a difference histogram, and the determination unit 34 determines the presence or absence of a background based on the difference histogram and the frequency relationship in each division in the difference histogram. Therefore, it is possible to determine the presence / absence of a highly accurate background in consideration of “color”.

  Further, by performing output based on the determination result of the presence / absence of the background, it is possible to obtain an output result with high legibility that also considers “color”.

  Further, based on the signal from the region separation processing unit 12, the difference described above is applied only to the pixels classified into the third region (photo region) except for the pixels classified into the character region and the dot region. It may be configured to perform density conversion processing based on a histogram.

  In this case, as shown in FIG. 6, the input gradation correction unit 13 is provided with a pixel selection unit (pixel selection unit) 50 that selects pixels classified into the third region. That is, only the pixels in the third region forming the background are set as the background determination target pixels. Further, the pixel selection unit 50 selects pixels in the third region based on the region separation signal from the region separation processing unit 12.

  Here, as a method of separating image data from characters, halftone dots, and third region (photo region), for example, the method described in “Image Electronics Society of Japan Proceedings 90-06-04” may be used. it can. This method will be described below.

First, the following determinations (a) to (c) are performed in a block of M × N (M and N are natural numbers) pixels centered on the target pixel.
(A) The average value D _AVE of the signal level is obtained for the central nine pixels in the block, and each pixel in the block is binarized using the average value. Further, the maximum pixel signal level D _MAX and the minimum pixel signal level D _MIN are also obtained at the same time.
(B) In the halftone dot area, the halftone dot area is identified by utilizing the fact that the fluctuation of the image signal in the small area is large and the density is higher than the background. For the binarized data, the number of change points from 0 to 1 and the number of change points from 1 to 0 are obtained in the main scanning and sub-scanning directions, and are set as K _H and K _V , respectively. If both of them exceed the threshold value compared with T _H and T _V , the halftone dot region is set. Further, in order to prevent erroneous determination with the background, D _MAX -D _AVE and D _AVE -D _MIN are compared with threshold values B ₁ and B ₂ , respectively. Specifically, when the following equations (2) to (5) are simultaneously established, a halftone dot region is set, and other cases are set as non-halftone dot regions.

K _H > T _H (2)
K _V > T _V (3)
D _MAX -D _AVE > B ₁ (4)
D _AVE −D _MIN > B ₂ (5)
(C) In the character area, since the difference between the maximum signal level D _MAX and the minimum signal level D _MIN is large and the density is high, the character area is identified as follows. The maximum signal level D _MAX and the minimum signal level D _MIN previously obtained in the non-halftone area and their difference D _SUB are compared with threshold values P _A , P _B and P _C , respectively, and the following equation (6 ) To (8), the character area is set, and the other case is set to the third area (photo area).

_{D MAX> P A ... (6} )
D _MIN <P _B (7)
_{D SUB> P C ... (8} )
In this way, the region separation is performed, and the image is separated into the character portion, the halftone dot photograph portion, and the third region (photo region including the background portion), and only the pixels related to the third region are subjected to difference histogram processing. As a result, it is possible to delete information on portions that are not backgrounds in advance, and the effect of improving the background detection accuracy and the processing speed of image processing can be obtained.

[Embodiment 2]
The following will describe another embodiment of the present invention with reference to FIGS. For convenience of explanation, members having the same functions as those shown in the drawings of the above-described embodiment are given the same reference numerals and explanation thereof is omitted.

  The color image processing apparatus according to the present embodiment is different from the color image processing apparatus according to the first embodiment only in the configuration of the input tone correction unit. That is, the configuration is the same as that of the color image processing apparatus of the first embodiment except for the configuration of the input tone correction unit. Therefore, hereinafter, the input tone correction unit will be described.

  As shown in FIG. 7, the input tone correction unit 13 ′ according to the present embodiment includes a maximum value calculation unit (maximum value calculation unit) 30, a minimum value calculation unit (minimum value calculation unit) 31, and a difference calculation unit (difference). Calculation means) 32, difference histogram creation section (difference histogram creation means) 33, determination section (background determination means) 34 ', conversion table selection section 35', density conversion section 36 ', C histogram creation section (color component histogram creation means) 37, an M histogram creation unit (color component histogram creation means) 38, and a Y histogram creation unit (color component histogram creation means) 39.

  The C histogram creation unit 37 creates a density histogram of single C (cyan) data (hereinafter referred to as a C histogram) after the complementary color inversion is performed by the complementary color inversion unit described in the first embodiment. Further, the M histogram creation unit 38 creates a density histogram (hereinafter referred to as an M histogram) of M (magenta) data alone. Furthermore, the Y histogram creating unit 39 creates a histogram of density of Y (yellow) data alone (hereinafter referred to as Y histogram). The histograms of the C histogram, M histogram, and Y histogram are also referred to as color component histograms (histograms relating to pixel values of color components).

  The determination unit 34 ′ uses the C histogram created by the C histogram creation unit 37, the M histogram created by the M histogram creation unit 38, and the Y histogram created by the Y histogram creation unit 39 according to the first embodiment. Are provided with CMY detection category calculation sections 40 for calculating detection categories (hereinafter referred to as detection categories for C, M, and Y), respectively. In addition, the determination unit 34 ′ includes an average value calculation unit 41 that calculates an average value of the values of the C, M, and Y detection sections (hereinafter referred to as an average value of the detection sections). Here, the CMY detection category calculation unit 40 and the average value calculation unit 41 are also collectively referred to as background level calculation means.

  In addition, when calculating the detection classification regarding C, M, and Y, the determination method of Patent Document 2 shown in the first embodiment is used for each histogram, and the first embodiment shows the same. Calculation is made without providing such an upper limit.

  Further, as in the first embodiment, the determination unit 34 ′ calculates a detection category related to the difference from the difference histogram created by the difference histogram creation unit 33. Then, it is determined whether or not there is a ground based on the detection classification regarding the difference and the detection classification regarding C, M, and Y. A specific method for determining whether or not there is a ground will be described later.

  Further, when the determination unit 34 ′ determines that there is a background, the determination unit 34 ′ sets the average value of the detection classification as the background level.

  The conversion table selection unit 35 ′ selects a conversion table according to the determination result of the presence / absence of the background of the determination unit 34 ′ and the background level when the determination unit 34 ′ determines that there is a background. .

  For example, when the determination unit 34 ′ determines that there is no background, the conversion table selection unit 35 ′ selects a through table indicated by * in FIG. . In other cases, as the background level increases, density conversion tables indicated by ◆, ■, ▲, and X in the figure are used. That is, as shown in the figure, a conversion table is selected in which the output value with respect to the input value is reduced within the predetermined input range (0 to 128 gradations in the figure) as the background level increases. In FIG. 8, the relationship between the input and the output in each density conversion table is shown by a graph, but the actual density conversion table is set by, for example, 8-bit data.

  Further, instead of using a plurality of density conversion tables as described above, only the basic density conversion table (through table when there is no background) is used. Alternatively, the inclination and the shift amount in the basic density conversion table may be calculated.

  The conversion table selection unit 35 ′ may be provided with a background level correction unit (background level correction unit) 49 as shown in FIG. The function of the background level correction unit 49 and the configuration using the background level correction unit 49 will be described later.

  The density conversion unit 36 ′ uses the conversion table selected by the conversion table selection unit 35 ′ to perform the complementary color inversion on the pixel image data after the complementary color inversion similar to the density conversion unit 36 of the first embodiment. Perform density conversion processing. Then, the data whose density has been converted by the density conversion unit 36 ′ is sent to the color correction unit 14.

  FIG. 9 is a flowchart showing an operation example of the input tone correction unit 13 '. In the present embodiment, the difference calculation unit 32 calculates the difference between the maximum value and the minimum value for each pixel based on the CMY data that has been subjected to complementary color inversion (S11). After S11, the difference histogram creation unit 33 creates a difference histogram based on the difference value, and each histogram creation unit (37, 38, 39) creates each histogram described above. (S12).

  After S12, the determination unit 34 'first determines from the difference histogram whether or not there is a background (S13). The determination in S13 may use the determination method of Patent Document 2 shown in the first embodiment.

  If it is determined in S13 that there is no background, a table without background removal is selected by the conversion table selection unit 35 '(S16). On the other hand, if it is determined in S13 that there is a background, the determination unit 34 'further compares the values of the respective detection sections relating to C, M, and Y with the threshold value Th1 (S14).

  In S14, when there is at least one detection section having a value larger than the threshold value Th1 among the detection section values related to C, M, and Y, the process proceeds to S16, and the conversion table selection unit 35 ′ uses the background removal no table (above). Through table) is selected. Even if it is determined that the “color tone” is small in the difference histogram, the background should not be removed if the color component density (pixel value) is determined to be high in the color component histogram. Because.

  On the other hand, in S14, when the values of the detection sections for C, M, and Y are all smaller than the threshold value Th1, the conversion table selection unit 35 ′ uses the average value of the detection sections (that is, the background level). A conversion table corresponding to is selected (S15).

  After S15 or S16, density conversion processing is performed by the density conversion unit 36 'using the conversion table selected in S15 or S16 (S17). Thus, a series of processing ends.

  Here, FIGS. 10 to 13 show specific examples of the processing according to FIG.

  FIG. 10 shows the above four types of histograms relating to image data of a certain color input image. More specifically, FIG. 10A shows a difference histogram created by the difference histogram creation unit 33 for the image data. FIG. 10B shows a C histogram created by the C histogram creating unit 37 for the image data. FIG. 10C shows the M histogram created by the M histogram creating unit 38 for the image data. FIG. 10D is a Y histogram created by the Y histogram creating unit 39 for the image data.

  Here, the determination method described in Patent Document 2 is used, and the threshold value Th1 is set to 11 in advance. Note that the upper limit value of the search in the difference histogram is set to category 5 as in the first embodiment. For the C, M, and Y histograms, no upper limit is provided as described above.

  First, in the difference histogram shown in FIG. 10A, it is determined by the above determination method that a predetermined condition in the determination method is satisfied in the section 3 that is a section smaller than the section 5 that is the upper limit value, and there is a background. Is done.

  In the C histogram shown in FIG. 10B, the predetermined condition is satisfied in the section 8 by the determination method, and the detection section for C is 8. Further, in the M histogram shown in FIG. 10C, the predetermined condition is satisfied in the category 5 by the determination method, and the detection category for M is 5. In the Y histogram shown in FIG. 10D, the predetermined condition is satisfied in the section 7 by the determination method, and the detection section related to Y is 7.

  Here, in any of the C, M, and Y histograms, the detection section is smaller than the threshold value Th1, and therefore the determination unit 34 'determines that there is a background. Then, the average value calculation unit 41 calculates that the average value of the detection sections regarding C, M, and Y is 7. In this calculation, rounding is performed and the average is 7. However, it is not limited to this.

  Further, when the average value of the detection section is calculated, the conversion table selection unit 35 'selects the density conversion table corresponding to the average value 7 (background level) of the detection section. Then, density conversion is performed by the density conversion unit 36 ′ using the selected density conversion table. Therefore, the background removal process can be performed with a more appropriate removal amount by referring to the information (density information) of each color component that cannot be detected by the difference histogram.

  FIG. 11 shows the above four types of histograms for image data of a certain monochrome input image. More specifically, FIG. 11A shows a difference histogram created by the difference histogram creation unit 33 for the image data. FIG. 11B shows a C histogram created by the C histogram creating unit 37 for the image data. FIG. 11C shows an M histogram created by the M histogram creating unit 38 for the image data. FIG. 11D shows a Y histogram created by the Y histogram creating unit 39 for the image data.

  First, in the difference histogram shown in FIG. 11A, it is determined by the above determination method that the predetermined condition in the determination method is satisfied in the section 1 that is a section smaller than the section 5 that is the upper limit value, and there is a ground. Is done.

  In the C histogram shown in FIG. 11B, the predetermined condition is satisfied in section 1 by the determination method, and the detection section for C is 1. Further, in the M histogram shown in FIG. 10C, the predetermined condition is satisfied in section 1 by the determination method, and the detection section related to M is 1. In the Y histogram shown in FIG. 10D, the predetermined condition is satisfied in section 1 by the determination method, and the detection section related to Y is 1.

  Here, in any of the C, M, and Y histograms, the detection section is smaller than the threshold value Th1, and therefore the determination unit 34 'determines that there is a background. Then, the average value calculation unit 41 calculates that the average of the detection classifications regarding C, M, and Y is 1.

  Further, when the average value of the detection section is calculated, the conversion table selection unit 35 'selects the density conversion table corresponding to the average value 1 (background level) of the detection section. Then, density conversion is performed by the density conversion unit 36 ′ using the selected density conversion table. Therefore, also in this case, the background removal processing can be performed with a more appropriate removal amount by referring to the information (density information) of each color component that cannot be detected by the difference histogram.

  FIG. 12 shows the four types of histograms relating to image data of other color input images. This other color input image is a light image in which three colors of C, M, and Y are present in the same area as shown in FIG.

  FIG. 12A shows a difference histogram created by the difference histogram creation unit 33 for the image data. FIG. 12B shows a C histogram created by the C histogram creating unit 37 for the image data. FIG. 12C shows an M histogram created by the M histogram creating unit 38 for the image data. FIG. 12D is a Y histogram created by the Y histogram creating unit 39 for the image data.

  First, in the difference histogram shown in FIG. 12A, it is determined by the determination method that the predetermined condition is not satisfied and the background is not present in a section smaller than the upper limit section 5. Further, the table without background removal is selected by the conversion table selection unit 35 '. Then, density conversion is performed by the density conversion unit 36 ′ using the selected background removal table.

  As described above, in the present embodiment, background detection is performed also using four types of histograms. From the difference histogram, “color” information for each pixel is obtained. Further, although information for each pixel cannot be obtained from the C, M, and Y histograms, density distribution information in a single color can be obtained. Therefore, it is possible to perform more optimal background removal processing by referring not only to the difference histogram but also to each color component histogram.

  In the present embodiment, the operation example of the input tone correction unit 13 'is shown in FIG. 9, but the present invention is not limited to this. Hereinafter, another operation example will be described based on the flowchart of FIG.

  The process shown in FIG. 14 is a method of determining the background level and the characteristics of the background and setting the level of background removal step by step. In this case, the background level correction unit (background level correction means) 49 described above is used. The background level correction unit 49 corrects the average value (background level) of the detection sections obtained from the determination unit 34 '.

  First, the difference calculation unit 32 calculates the difference between the maximum value and the minimum value for each pixel (S21). After S21, the difference histogram creation unit 33 creates a difference histogram based on the difference value, and each histogram creation unit (37, 38, 39) creates each color component histogram described above. (S22).

  After S22, the determination unit 34 'first determines whether or not there is a background from the difference histogram (S23). The determination in S23 uses the determination method of Patent Document 2 shown in the first embodiment.

  If it is determined in S23 that there is no background, the conversion table selection unit 35 'selects a table without background removal as the density conversion table (S30). On the other hand, when it is determined in S23 that there is a background, the determination unit 34 'determines whether or not the value of the detection classification regarding the difference is larger than the threshold Th2 (S24).

  If it is determined in S24 that the value of the detection section related to the difference is less than the threshold Th2, the conversion table selection unit 35 ′ selects a density conversion table corresponding to the average value of the detection section (S28). . That is, in this case, the background level correction unit 49 selects the density conversion table corresponding to the average value of the detection section without correcting the average value of the detection section.

  On the other hand, if it is determined in S24 that the value of the detection classification related to the difference is larger than the threshold value Th2, the determination unit 34 ′ further makes the minimum value larger than the threshold value Th3 and the minimum difference is larger than the threshold value Th4. Is determined (S25). The minimum value refers to the value of the smallest detection section among the detection section values related to C, M, and Y. That is, when the value of each detection category is LevC, LevM, and LevY, it means Min (LevC, LevM, LevY). The minimum difference is Min (| LevC−LevM |, | LevM−LevY |, | LevY−LevC |).

  When it is determined in S25 that the minimum value is equal to or less than the threshold Th3 and / or the minimum difference is equal to or less than the threshold Th4, the conversion table selection unit 35 ′ determines the average value (background level) of the detection classification. A density conversion table that provides a slightly smaller background removal amount than the density conversion table is selected (S29). That is, the background level correction unit 49 corrects the average value (background level) of the detection sections so that the background removal amount is reduced, and a density conversion table corresponding to the corrected value is selected.

  In this case, since both the minimum value and the minimum difference are small values, the background is a primary color or a secondary color that does not include a gray component. In other words, even if the background is removed, the color change hardly occurs. It can be judged. For this reason, the density conversion in this case is such that the background removal amount is more conservative than the background removal amount based on the background level before correction, and the show-through is obtained.

  On the other hand, in S25, when it is determined that the minimum value is larger than the threshold Th3 and the minimum difference is larger than the threshold Th4, the determination unit 34 ′ determines the detection classification (LevC, LevM, LevY) for C, M, and Y. ) Is determined as to whether LevC <LevM <LevY (S26).

  If the relationship of LevC <LevM <LevY is established in S26, the conversion table selection unit 35 'determines that the paper quality is poor, such as newspapers, and further yellowing due to fading, and the average value (background level) The density conversion table that has a larger background removal amount than the density conversion table corresponding to () is selected (S27). That is, the background level correction unit 49 corrects the average value (background level) of the detection sections so that the background removal amount increases, and a density conversion table corresponding to the corrected value is selected.

  On the other hand, if the relationship of LevC <LevM <LevY is not satisfied in S26, the conversion table selection unit 35 ′ determines that the background is not a background that does not change color due to the background removal, and is not a poor paper quality or yellowish background. A table without background removal is selected as the density conversion table (S30). That is, in this case, the correction of the average value of the detection section is not performed.

  Then, after any of the processes of S27 to 29 is completed, the density conversion process is performed by the density conversion unit 36 'using the density conversion table selected in each of these processes (S31). Thus, a series of processing ends.

  As described above, it is possible to perform background removal processing with a more optimal removal amount by determining the change in color, the presence or absence of fading, and the like.

  In addition, the degree of correction of the background level may be changed according to the minimum value or the value of the minimum difference (degree of change in color). For example, as both the minimum value and the minimum difference become smaller, the degree of correction is also changed so that the background removal amount is close to the normal removal amount (removal amount without correction).

  By the way, as shown in FIG. 15, the determination unit 34 ′ includes a shape determination unit (shape determination unit) 42 that determines the shape of the histogram, and the conversion table selection unit 35 ′ further includes a background level correction unit (background). (Level correction means) 49 may be provided. Hereinafter, the case where the shape determination unit 42 and the background level correction unit 49 are provided will be described with reference to FIGS. 15 and 16.

  When the show-through has occurred, the difference value for the portion where the show-through has occurred tends to be substantially the same value as the difference value in the portion where the show-through has not occurred. However, the value of each color component (C, M, Y) for the portion where the show-through occurs is not the same value as each color component in the portion where the show-through does not occur, and the value of each color component becomes high. Has a tendency.

  Therefore, when an image showing show-through is input, the shape of the histogram related to the difference is sharp in the vicinity of the detection section related to the difference, whereas the shape of the histogram related to the color component is close to the detection section related to the color component. It tends to be broader on the higher density side (the higher pixel value side). In view of this characteristic, a density conversion table is selected and density conversion processing is performed.

  FIG. 16 is a flowchart illustrating an operation example of the input tone correction unit 13 ′ including the shape determination unit 42 and the background level correction unit 49. First, based on the CMY data that has undergone complementary color inversion, the difference calculation unit 32 calculates the difference between the maximum value and the minimum value for each pixel (S41). After S41, the difference histogram creation unit 33 creates a difference histogram based on the difference value, and each histogram creation unit (37, 38, 39) creates each color component histogram described above. (S42).

  After S42, the determination unit 34 'first determines from the difference histogram whether or not there is a background (S43). The determination in S43 may use the determination method of Patent Document 2 shown in the first embodiment.

  If it is determined in S43 that there is no background, a table without background removal is selected by the conversion table selection unit 35 '(S48). On the other hand, if it is determined in S43 that there is a background, the determination unit 34 'determines whether there is a background from each of the color component histograms (S44).

  If it is determined in S44 that there is no background, a table without background removal is selected by the conversion table selection unit 35 '(S48). On the other hand, if it is determined in S44 that there is a background, the shape determining unit 42 has a sharp shape with a difference histogram of a predetermined level or higher and each color component histogram has a shape of a predetermined level or higher. It is determined whether or not the shape is broad (S45).

  In S45, when the shape of the difference histogram is a sharp shape of a predetermined level or higher and the shape of each color component histogram is a broad shape of a predetermined level or higher, the conversion table selection unit 35 ′ causes C, A density conversion table having a larger background removal amount than the density conversion table corresponding to the average value of the detection classification (background level) values for M and Y is selected (S46). That is, the background level correction unit 49 corrects the average value (background level) of the detection sections so that the background removal amount increases, and a density conversion table corresponding to the corrected value is selected.

  On the other hand, in S45, if the shape of the difference histogram is not a sharp shape of a predetermined level or higher and / or the shape of each color component histogram is not a broad shape of a predetermined level or higher, the conversion table selection unit 35 ′ A density conversion table corresponding to the average value of the detection classification (background level) values for C, M, and Y is selected (S47). That is, in this case, the background level correction unit 49 selects the density conversion table corresponding to the average value of the detection section without correcting the average value of the detection section.

  When any of the processes from S46 to S48 is completed, the density conversion process is performed by the density conversion unit 36 'using the density conversion table selected in each process (S49).

  Note that an example of a specific method for determining the shape will be described in Embodiment 3 to be described later with reference to FIG. However, in Embodiment 3, the case where luminance is used instead of color components is described, but the method itself is the same.

  As described above, the presence / absence of show-through is determined by the shape determination unit 42, and the average value (background level) of the detection classification is corrected by the background level correction unit 49 according to the determination result. A removal process can be performed.

  Further, based on the signal from the region separation processing unit 12, the difference described above is applied only to the pixels classified into the third region (photo region) except for the pixels classified into the character region and the halftone dot region. It may be configured to perform density conversion processing based on the histogram, C histogram, M histogram, and Y histogram.

  Also in this case, as shown in FIG. 17, as in the first embodiment (see FIG. 6), the input tone correction unit 13 ′ selects a pixel selection unit (pixel selection unit) that selects the pixels classified into the third region. Means) 50 may be provided, and only the pixels in the third region forming the background may be set as the background determination target pixels.

  Also in this case, as described in the first embodiment, it is possible to delete information on a portion that is not a background in advance, and an effect of improving the background detection accuracy and the processing speed of image processing can be obtained.

[Embodiment 3]
The following will describe another embodiment of the present invention with reference to FIGS. For convenience of explanation, members having the same functions as those shown in the drawings of Embodiments 1 and 2 are given the same reference numerals, and descriptions thereof are omitted.

  The color image processing apparatus according to the present embodiment is different from the color image processing apparatus according to the first embodiment only in the configuration of the input tone correction unit. That is, the configuration is the same as that of the color image processing apparatus of the first embodiment except for the configuration of the input tone correction unit. Therefore, hereinafter, the input tone correction unit will be described as in the second embodiment.

  As shown in FIG. 18, the input tone correction unit 13 ″ according to the present embodiment includes a maximum value calculation unit 30, a minimum value calculation unit 31, a difference calculation unit 32, a difference histogram creation unit 33, a determination unit (background determination unit). ) 34 ″, a conversion table selection unit 35 ″, a density conversion unit 36 ″, a luminance value calculation unit (luminance value calculation unit) 43, and a luminance histogram generation unit (luminance histogram generation unit) 44.

  The luminance value calculation unit 43 creates luminance data L from the CMY signal after the complementary color inversion is performed by the complementary color inversion unit described in the first embodiment. The luminance data L is calculated by the following equation (9) using, for example, C, M, and Y data.

L = a · C + b · M + c · Y (9)
In general, a, b, and c in the above equation (9) use values (a: b: c = 1.000: 4.5907: 0.0601) that are not in the same ratio for converting C, M, and Y data into luminance data. It is done. When the above values are used for a, b, and c, data that is almost faithful to the luminance of the actual image can be obtained. Further, by setting a = b = c (= 1/3), it is possible to grasp the characteristics of the document (characteristics related to the color), and to set a ratio that can detect a more colored background. it can.

  The luminance data L of the above equation (9) is obtained based on the C, M, and Y data, but the luminance data based on the R, G, and B data without performing complementary color inversion. L ′ may be used. In the following, description will be made using luminance data L using C, M, and Y data.

  The luminance histogram creation unit 44 creates a luminance histogram based on the luminance data L calculated by the luminance value calculation unit 43.

  The determination unit 34 ″ calculates a detection category (hereinafter referred to as a detection category related to luminance) from the luminance histogram created by the luminance histogram creation unit 44 in the same manner as the detection category related to difference in the first embodiment. And a luminance detection category calculation unit (background level calculation means) 45. Here, the detection category relating to luminance uses the determination method of Patent Document 2 shown in the first embodiment for the luminance histogram. However, in this case, the above upper limit value is set separately.

  Further, as in the first embodiment, the determination unit 34 ″ calculates the detection section related to the above-described difference from the difference histogram created by the difference histogram creation unit 33. Then, the detection classification related to the difference and the detection classification related to the luminance Based on the above, it is determined whether or not there is a background, and a specific method for determining whether or not there is a background will be described later.

  Furthermore, when the determination unit 34 ″ determines that there is a background, the determination unit 34 ″ sets the value of the detection category related to the luminance as the background level.

  The conversion table selection unit 35 ″ selects the conversion table according to the background level when the determination unit 34 ″ determines that there is a background, and when the determination unit 34 ″ determines that there is a background. .

  For example, as in the first embodiment, when the determination unit 34 ″ determines that there is no background, the conversion table selection unit 35 ″ indicates, as shown in FIG. Select the table for the indicated thru. Other than this, as the background level increases, the density conversion tables indicated by ◆, ■, ▲, and X in the figure may be used.

  The conversion table selection unit 35 ″ may be provided with a background level correction unit (background level correction means) 49 ′ as shown in the figure. The function of the background level correction unit 49 ′, and this The configuration using the ground level correction unit 49 ′ will be described later.

  The density conversion unit 36 ″ performs density conversion processing using the conversion table selected by the conversion table selection unit 35 ″. Then, the data subjected to the density conversion by the density conversion unit 36 ″ is sent to the color correction unit 14.

  FIG. 19 is a flowchart showing an operation example of the input tone correction unit 13 ″. In this embodiment, the difference calculation unit 32 performs the above maximum value for each pixel based on CMY data that has been subjected to complementary color inversion. (S51) After S51, the difference histogram creation unit 33 creates a difference histogram based on the difference value, and the luminance histogram creation unit 44 The luminance histogram described above is created (S52).

  After S52, the determination unit 34 ″ first determines whether or not there is a background from the difference histogram (S53). The determination in S53 is the patent document 2 shown in the first embodiment. If it is determined in S53 that there is no background, a table without background removal is selected by the conversion table selection unit 35 ″ (S56).

  On the other hand, if it is determined in S53 that there is a background, the determination unit 34 ″ further determines whether or not there is a background from the luminance histogram (S54).

  If it is determined in S54 that there is no background, a table without background removal is selected by the conversion table selection unit 35 '' (S56). On the other hand, if it is determined in S54 that there is a background, conversion is performed. The table selection unit 35 ″ selects a density conversion table corresponding to the value (background level) of the detection category relating to the luminance (S55).

  After S55 or S56, density conversion processing is performed using the density conversion table selected for each (S57). As a result, a series of processing ends.

  Here, FIGS. 20 to 22 show specific examples of the processing according to FIG.

  FIG. 20 shows the two types of histograms relating to image data of a certain color input image. More specifically, FIG. 20A shows a difference histogram created by the difference histogram creation unit 33 for the image data. FIG. 10B is a luminance histogram created by the luminance histogram creation unit 44 for the image data.

  Note that the upper limit value of the search in the difference histogram is set to category 5 as in the first embodiment. Further, the upper limit value of the search in the luminance histogram is set to 11. However, these upper limit values are not limited to these values.

  First, in the difference histogram shown in FIG. 20A, it is determined by the above determination method that a predetermined condition in the determination method is satisfied in the section 3 that is a section smaller than the section 5 that is the upper limit value, and there is a background. Is done.

  In the luminance histogram shown in FIG. 20B, the predetermined condition is satisfied in Section 6 by the determination method. Therefore, a predetermined condition in the determination method is satisfied in a section smaller than the section 11 that is the upper limit value for luminance, and it is determined that there is a background.

  Accordingly, the conversion table selection unit 35 ″ selects a density conversion table corresponding to the value 6 (background level) of the detection category relating to luminance. Then, using the selected density conversion table, the density conversion unit 36 ″. Thus, density conversion is performed. Therefore, the background removal processing can be performed with a more appropriate removal amount by considering not only the difference histogram but also the luminance value.

  FIG. 21 shows the two types of histograms relating to image data of a certain monochrome input image. More specifically, FIG. 21A shows a difference histogram created by the difference histogram creation unit 33 for the image data. FIG. 21B is a brightness histogram created by the brightness histogram creation unit 44 for the image data.

  Here, similarly to the above, the upper limit value of the search in the difference histogram is set as category 5, and the upper limit value of the search in the luminance histogram is set as category 11.

  First, in the difference histogram shown in FIG. 21 (a), it is determined by the above-described determination method that a predetermined condition in the determination method is satisfied in the section 1 that is a section smaller than the section 5 that is the upper limit value, and there is a background. Is done.

  In the luminance histogram shown in FIG. 21 (b), the predetermined condition is satisfied in Category 1 by the determination method. Therefore, a predetermined condition in the determination method is satisfied in a section smaller than the section 11 that is the upper limit value for luminance, and it is determined that there is a background.

  Accordingly, the conversion table selection unit 35 ″ selects a density conversion table corresponding to the value 1 (background level) of the detection category relating to luminance. Then, using the selected density conversion table, the density conversion unit 36 ″. Thus, density conversion is performed. Therefore, also in this case, the background removal process can be performed with a more appropriate removal amount by considering not only the difference histogram but also the luminance value.

  FIG. 22 shows the two types of histograms relating to the image data of the other color input images shown in FIG. More specifically, FIG. 22A shows a difference histogram created by the difference histogram creation unit 33 for the image data. FIG. 22B is a luminance histogram created by the luminance histogram creation unit 44 for the image data.

  Here, similarly to the above, the upper limit value of the search in the difference histogram is set as category 5, and the upper limit value of the search in the luminance histogram is set as category 11.

  First, in the difference histogram shown in FIG. 22A, it is determined by the determination method that the predetermined condition is not satisfied and the background is not present in a section smaller than the upper limit section 5. In addition, a table without background removal is selected by the conversion table selection unit 35 ″. Then, density conversion is performed by the density conversion unit 36 ″ using the selected table without background removal.

  As described above, in the present embodiment, background detection is performed also using the luminance histogram. From the difference histogram, “color” information for each pixel is obtained. Further, from the luminance histogram, information for each pixel cannot be obtained, but luminance (density) distribution information can be obtained. Therefore, by referring not only to the difference histogram but also to the luminance histogram, it is possible to perform more optimal background removal processing.

  Also in the present embodiment, it is possible to apply the method of determining the background level and the characteristics of the background and setting the level of background removal step by step as shown in FIG. 14 of the second embodiment. is there. FIG. 23 shows a flowchart in this case. In this case, the background level correction unit (background level correction means) 49 'described above is used. The background level correction unit 49 ′ corrects the detection classification (background level) obtained from the determination unit 34 ″.

  First, the difference calculation unit 32 calculates the difference between the maximum value and the minimum value for each pixel (S61). After S61, the difference histogram creation unit 33 creates a difference histogram based on the difference value, and the brightness histogram creation unit 44 creates the above-described brightness histogram (S62).

  After S62, the determination unit 34 ″ first determines whether or not there is a background from the difference histogram (S63). The determination in S63 is the patent document 2 shown in the first embodiment. The determination method is used.

  If it is determined in S63 that there is no background, the conversion table selection unit 35 ″ selects a table without background removal as the density conversion table (S68). On the other hand, in S63, it is determined that there is a background. In this case, the determination unit 34 ″ determines whether or not the value of the detection category regarding the difference is larger than the threshold value Th2 (S64).

  When it is determined in S64 that the value of the detection section related to the difference is less than the threshold Th2, the conversion table selection unit 35 ″ selects a density conversion table corresponding to the value of the detection section related to the brightness (S67). In other words, in this case, the density conversion table corresponding to the detection section is selected without correcting the detection section in the background level correction unit 49.

  On the other hand, if it is determined in S64 that the value of the detection category related to the difference is larger than the threshold Th2, the determination unit 34 ″ further determines whether or not the value of the detection category related to the luminance is equal to or greater than the threshold Th5. (S65).

  If it is determined in S65 that the detection category related to luminance is less than the threshold Th5, the conversion table selection unit 35 ″ selects a density conversion table corresponding to the value (background level) of the detection category related to luminance (S67). On the other hand, if it is determined in S65 that the detection category related to luminance is equal to or greater than the threshold Th5, the conversion table selection unit 35 ″ selects a table with a larger removal amount than the density conversion table corresponding to the value of the detection category related to luminance. (S66). That is, the background level correction unit 49 ′ corrects the detection section (background level) so that the background removal amount increases, and a density conversion table corresponding to the corrected value is selected.

  After any of the processes from S66 to S68 is completed, the density conversion process is performed by the density conversion unit 36 ″ using the density conversion table selected in each of these processes (S69). Thus, a series of processing ends.

  As described above, when the luminance is low, the background removal processing can be performed with a more optimal removal amount by further increasing the background removal amount by correction.

  As shown in FIG. 24, the determination unit 34 ″ includes a shape determination unit (shape determination unit) 42 ′ that determines the shape of the histogram, and the conversion table selection unit 35 ″ further includes a background level correction unit ( It may be configured to include a ground level correction means) 49 ′. Hereinafter, the case where the shape determination unit 42 ′ and the background level correction unit 49 ′ are provided will be described with reference to FIGS. 24 and 25.

  When the show-through has occurred, the difference value for the portion where the show-through has occurred tends to be substantially the same value as the difference value in the portion where the show-through has not occurred. However, the luminance value for the portion where the show-through has occurred is not the same as the luminance value in the portion where the show-through has not occurred, and the luminance value tends to be low.

  Therefore, when an image with show-through is input, the shape of the histogram related to the difference is sharp in the vicinity of the detection section related to the difference, whereas the shape of the histogram related to the brightness is low in the vicinity of the detection section related to the brightness. Tend to be broad on the side. In view of this characteristic, a density conversion table is selected and density conversion processing is performed.

  FIG. 25 is a flowchart showing an operation example of the input tone correction unit 13 ″ including the shape determination unit 42 ′ and the background level correction unit 49 ′. First, the difference is calculated based on the CMY data having the complementary colors inverted. A difference between the maximum value and the minimum value is calculated for each pixel by the calculation unit 32 (S71) After S71, a difference histogram creation unit 33 creates a difference histogram based on the difference value. At the same time, the luminance histogram creating unit 44 creates the above-described luminance histogram (S72).

  After S72, the determination unit 34 ″ first determines whether or not there is a background from the difference histogram (S73). The determination in S73 is the patent document 2 shown in the first embodiment. This determination method may be used.

  If it is determined in S73 that there is no background, a table without background removal is selected by the conversion table selection unit 35 ″ (S78). On the other hand, if it is determined in S73 that there is a background, determination is performed. The unit 34 ″ determines whether or not there is a background from the luminance histogram (S74).

  If it is determined in S74 that there is no background, a table without background removal is selected by the conversion table selection unit 35 ″ (S78). On the other hand, if it is determined in S74 that there is a background, the shape is determined. The determination unit 42 ′ determines whether or not the shape of the difference histogram is a sharp shape having a predetermined level or higher and the luminance histogram has a broad shape having a predetermined level or higher (S75).

  In S75, if the shape of the difference histogram is a sharp shape of a predetermined level or higher and the shape of the luminance histogram is a broad shape of a predetermined level or higher, the conversion table selection unit 35 ″ detects the luminance. A density conversion table having a background removal amount larger than the density conversion table corresponding to the classification (background level) is selected (S76), that is, the background level correction unit 49 ′ detects a classification ( (Background level) is corrected, and a density conversion table corresponding to the corrected value is selected.

  On the other hand, in S75, when the shape of the difference histogram is not a sharp shape of a predetermined level or higher and / or when the shape of the luminance histogram is not a broad shape of a predetermined level or higher, the conversion table selection unit 35 " A density conversion table corresponding to the detection category (background level) is selected (S77), that is, in this case, the detection level is not corrected by the background level correction unit 49 ′, and the average value of the detection category is used. The selected density conversion table is selected.

  Then, when any of the processes from S76 to S78 is completed, the density conversion process is performed by the density conversion unit 36 ″ using the density conversion table selected in each process (S79).

  As described above, the presence / absence of show-through is determined by the shape determination unit 42 ′, and the average value (background level) of the detection classification is corrected by the background level correction unit 49 ′ in accordance with the determination result. The background removal process can be performed.

  Here, an example of a specific method of the shape determination will be described. FIG. 26 is an example of a histogram. More specifically, FIG. 26A is a difference histogram, and FIG. 26B is a luminance histogram. In addition, the detection classification related to the difference obtained from the difference histogram is 3 (shaded portion in FIG. 26A), and the detection classification related to the luminance calculated from the luminance histogram is 3 (shaded portion in FIG. 26B).

  In the shape determination, the shape determination unit 42 ′ determines the number of divisions having a higher frequency than the frequency for each detection division, including the detection division.

  As shown in FIG. 26A, in the difference histogram, when the detection section 3 is included, the sections having a frequency greater than the frequency of the detection section 3 are two sections, that is, the section 3 and the section 5. In addition, as shown in FIG. 26B, in the luminance histogram, when the detection section 3 is included, the sections having a frequency greater than the frequency of the detection section 3 are five of the sections 3 to 7.

  Here, for example, if the threshold value of whether the shape is sharp or broad is set to 3, it can be determined that the difference histogram is a sharp shape and the luminance histogram is a broad shape. The shape determination method is not limited to the above.

  Further, based on the signal from the region separation processing unit 12, the difference described above is applied only to the pixels classified into the third region (photo region) except for the pixels classified into the character region and the halftone dot region. It may be configured to perform density conversion processing based on a histogram and a luminance histogram.

  Also in this case, as shown in FIG. 27, as in the first and second embodiments (see FIGS. 6 and 15), the pixel selection for selecting the pixels classified into the third region is performed in the input tone correction unit 13 ″. The unit (pixel selection means) 50 may be provided, and only the pixels in the third region forming the background may be set as the background determination target pixels.

  Also in this case, as described in the first and second embodiments, it is possible to delete information on a portion that is not a background in advance, and an effect of improving the background detection accuracy and the processing speed of image processing can be obtained.

  In each of the above first to third embodiments, the maximum value calculation unit 30 calculates the maximum value for each pixel among the pixel values of each color component given to each pixel. Further, the minimum value calculation unit 31 calculates the minimum value for each pixel among the pixel values of each color component given to each pixel. And the difference calculation part 32 calculated the difference of the said maximum value and the said minimum value in the same pixel for every pixel. Specifically, when there are three color components C, M, and Y, the difference is obtained for each pixel by max (C, M, Y) -min (C, M, Y). However, the present invention is not limited to this, and may be configured as follows.

  First, the maximum value calculation unit obtains an absolute value of a difference between pixel values of different color components among pixel values of each color component given to each pixel, and calculates the maximum value of the absolute values for each pixel. . Further, the absolute value of the difference between the pixel values of different color components among the pixel values of each color component given to each pixel is obtained by the minimum value calculation unit, and the minimum value of the absolute values is calculated for each pixel. . Then, the difference calculation unit calculates the difference between the maximum value and the minimum value in the same pixel for each pixel.

  Specifically, when there are three color components C, M, and Y, max {| CM |, | MY |, | YC |} -min {| CM |, | MY |, | YC |} The above difference may be obtained for each pixel.

  Also with the above configuration, it is possible to determine the presence / absence of a groundwork with high accuracy in consideration of “color”. The processing in the difference histogram creation unit 33 is the same as the processing shown in each of the above embodiments.

  Further, the following configuration may be used.

  First, the maximum value calculation unit obtains an absolute value of a difference between pixel values of different color components among pixel values of each color component given to each pixel, and calculates the maximum value of the absolute values for each pixel. . Specifically, when there are three color components C, M, and Y, max {| C-M |, | M-Y |, | Y-C |} is obtained for each pixel.

  Then, instead of the difference histogram 33, a maximum value histogram creating unit (maximum value histogram creating means) for creating a histogram relating to the maximum value based on the maximum value is provided. Furthermore, the background determination unit is configured to determine the presence or absence of a background based on a histogram relating to the maximum value.

  Also with this configuration, it is possible to determine the presence / absence of a base with high accuracy in consideration of “color”.

  Alternatively, a histogram relating to the maximum value may be created based on the maximum value among the pixel values of each color component given to each pixel, and the presence / absence of the background may be determined based on the histogram relating to the maximum value.

  Here, the magnitude of “color” is determined by the magnitude of the maximum value when attention is paid to a certain pixel. More specifically, “color” is larger as the maximum value is larger, and “color” is smaller as the maximum value is smaller. Therefore, considering the entire image, the background should be reproduced when the number of pixels having a large maximum value is large, and the background should be removed when the number of pixels having a small maximum value is large.

  Therefore, after creating a histogram related to the maximum value using the maximum value histogram creation means, the background judgment means determines whether or not there is a background based on the maximum value category and the frequency relationship in each category in the maximum value histogram. To do. This also brings about an effect that it is possible to determine the presence / absence of a highly accurate background in consideration of “color”.

  In addition, as described above, the image processing apparatus according to the present invention provides at least each color given to each pixel in the image processing apparatus that determines at least the presence or absence of the background of the document from the image data of each pixel obtained by reading the document. The absolute value of the difference between pixel values of different color components among the pixel values of the components is obtained, and the maximum value calculating means for calculating the maximum value of the absolute values for each pixel, and the maximum value based on the maximum value It can be said that the apparatus includes a maximum value histogram creating unit that creates a histogram relating to a value, and a background determination unit that determines the presence or absence of a background based on the histogram related to the maximum value.

  By the way, the ground includes a ground that should be reproduced and a ground that should be removed without being reproduced. That is, in the case of paper with “color” that is intentionally colored like color paper, the background should be reproduced, while the background with poor paper quality such as newspaper, fading, yellowing, The gray background that hinders readability should be removed.

  Here, the magnitude of “color” is determined by the magnitude of the maximum value when attention is paid to a certain pixel. More specifically, “color” is larger as the maximum value is larger, and “color” is smaller as the maximum value is smaller. Therefore, considering the entire image, the background should be reproduced when the number of pixels having a large maximum value is large, and the background should be removed when the number of pixels having a small maximum value is large.

  Therefore, after creating a histogram related to the maximum value using the maximum value histogram creation means, the background judgment means determines whether or not there is a background based on the maximum value category and the frequency relationship in each category in the maximum value histogram. To do.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  In addition, by performing output based on the determination result of the presence or absence of the background, an output result with high legibility can be obtained in consideration of “color”.

  In addition, as described above, the image processing apparatus according to the present invention provides at least each color given to each pixel in the image processing apparatus that determines at least the presence or absence of the background of the document from the image data of each pixel obtained by reading the document. An absolute value of a difference between pixel values of different color components among the pixel values of the components is obtained, a maximum value calculating means for calculating the maximum value of the absolute values for each pixel, and each color component given to each pixel A minimum value calculating means for calculating an absolute value of a difference between pixel values of different color components among the pixel values, and calculating a minimum value of the absolute values for each pixel; and the maximum value and the minimum value of the same pixel A difference calculation means for calculating a difference for each pixel, a difference histogram creation means for creating a histogram relating to the difference based on the difference, and a histogram based on the histogram relating to the difference. It can be said that the configuration and a base determination means for determining whether a.

  By the way, the ground includes a ground that should be reproduced and a ground that should be removed without being reproduced. That is, in the case of paper with “color” that is intentionally colored like color paper, the background should be reproduced, while the background with poor paper quality such as newspaper, fading, yellowing, The gray background that hinders readability should be removed.

  Here, the magnitude of “color” is determined by the difference between the maximum value and the minimum value when attention is paid to a certain pixel. More specifically, the larger the difference, the larger the “color”, and the smaller the difference, the “color” becomes smaller. Therefore, considering the entire image, the background should be reproduced when the number of pixels with a large difference is large, and the background should be removed when the number of pixels with a small difference is large.

  Therefore, first, a difference between the maximum value calculated by the maximum value calculating means and the minimum value calculated by the minimum value calculating means is calculated by the difference calculating means. Then, after creating a histogram relating to the difference by the difference histogram creating means, the background determining means determines the presence / absence of the background based on the difference category and the frequency relationship in each section in the difference histogram.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with high accuracy in consideration of “color”.

  In addition, by performing output based on the determination result of the presence or absence of the background, an output result with high legibility can be obtained in consideration of “color”.

  In addition, the image processing apparatus according to the present invention includes a color component histogram creating unit that creates a histogram related to the pixel value of each color component based on the pixel value of each color component of each pixel, and the pixel of each color component A background level calculation unit that calculates a background level based on a histogram relating to the value, and when the background determination unit determines that there is a background based on the histogram related to the difference, the background level calculated by the background level calculation unit It can be said that the background removal processing is performed based on the level.

  Here, the background level is calculated by the background level calculation means based on the histogram relating to the pixel value of each color component. When the background determination unit determines that there is a background based on the histogram relating to the difference, a background removal process is performed based on the background level calculated by the background level calculation unit.

  Therefore, when the background determination unit determines that there is a background, the background removal processing is performed based on the background level in consideration of the histogram regarding the pixel value of each color component.

  Therefore, it is possible to perform background removal processing with an optimum background removal amount in consideration of information on each color component.

  Further, in the image processing apparatus according to the present invention, the background determination unit further determines the presence / absence of a background based on a histogram related to the pixel value of each color component, and the background determination unit generates a histogram related to the pixel value of each color component. If it is determined that there is a background, it can be said that the background removal processing is performed based on the background level calculated by the background level calculation means.

  Here, even if it is determined that the “color tone” is small based on the histogram relating to the difference, in the histogram relating to the pixel value of the color component, if the frequency is high in the category where the pixel value is large, the background is applied. Should be reproduced.

  Therefore, there is an effect that the background determination unit can determine the presence / absence of the background with higher accuracy by further determining the presence / absence of the background based on the histogram relating to the pixel values of the respective color components.

  In the image processing apparatus according to the present invention, the presence / absence of a background to be removed is determined by a histogram section related to the difference that satisfies a predetermined condition, and the section has a larger difference than the predetermined section. In this case, it can be said that the background level correcting unit corrects the background level calculated by the background level calculating unit.

  Here, when the category is a category having a larger difference than the predetermined category, that is, when the “color tone” of the read document is somewhat large, the calculation is performed by the same method as when the “color tone” is small. For example, it is preferable to perform the background removal processing in consideration of the degree of change in “color” due to the background removal processing, the presence / absence of fading of the original, and the like, rather than performing the background removal processing at the background level.

  Therefore, by correcting the background level by the background level correction means, it is possible to perform the background removal process with a more appropriate removal amount.

  Further, in the image processing apparatus according to the present invention, the background determination unit further performs background removal processing based on the background level according to the value of each section of the histogram related to the pixel value of each color component that satisfies a predetermined condition. In addition, the degree of change in color when performing the above is determined, and the background level correction means corrects the background level according to the degree of change.

  Therefore, for example, when the degree of change in color is small, the background level can be corrected in accordance with the degree of change in color, such as correcting the background level so as to reduce the amount of background removal.

  Therefore, it is possible to perform the background removal process with the removal amount in consideration of the degree of change in color.

  Further, in the image processing apparatus according to the present invention, the background determination unit further determines the presence / absence of a fading based on the value of each section, and the background level correction unit determines the degree of change and the presence / absence of a fading. It can be said that the background level is corrected accordingly.

  Therefore, the ground level can be corrected according to the degree of change and the presence or absence of fading.

  Therefore, it is possible to perform the background removal process with a removal amount in consideration of the degree of change in color and the presence or absence of fading.

  Further, the image processing apparatus according to the present invention includes a luminance value calculating unit that calculates a luminance value of each pixel, a luminance histogram generating unit that generates a histogram relating to a luminance value based on the luminance value of each pixel, and the luminance It can be said that the apparatus includes a background level calculation unit that calculates the background level based on the histogram.

  Here, the background level is calculated by the background level calculation means based on the histogram relating to the luminance value. When the background determination unit determines that there is a background based on the histogram relating to the difference, a background removal process is performed based on the background level calculated by the background level calculation unit.

  Therefore, when the background determination unit determines that there is a background, the background removal processing is performed based on the background level in consideration of the histogram relating to the luminance value.

  Therefore, it is possible to perform background removal processing with an optimum background removal amount in consideration of luminance information.

  Further, in the image processing apparatus according to the present invention, the background determination unit further determines the presence / absence of a background based on a histogram related to luminance values, and the background determination unit determines that there is a background based on a histogram related to luminance values. In this case, it can be said that the background removal processing is performed based on the background level calculated by the background level calculation means.

  Here, when it is determined that “color tone” is small based on the histogram relating to the difference, it is possible to set a more appropriate removal amount in consideration of the luminance information from the detection result in the histogram relating to the luminance value. Become.

  Therefore, there is an effect that it is possible to determine the presence / absence of the background with higher accuracy by the background determination means further determining the presence / absence of the background based on the histogram relating to the luminance value.

  In the image processing apparatus according to the present invention, the presence / absence of a background to be removed is determined by a histogram section related to the difference that satisfies a predetermined condition, and the section has a larger difference than the predetermined section. And if the histogram section relating to the luminance value satisfying a predetermined condition is a higher-class section than the predetermined section, the base level calculated by the base level calculating means is used. It can be said that the configuration includes a ground level correction means for correcting.

  Here, when the category of the histogram relating to the luminance value that satisfies the predetermined condition is a category having a larger difference than the predetermined category, that is, when the “color” of the read document is somewhat large, It is preferable to perform background removal by increasing the amount of background removal, rather than performing background removal processing at the background level calculated by the same method as when “color” is small.

  Therefore, by correcting the background level by the background level correction means, it is possible to perform the background removal process with a more appropriate removal amount.

  In the present invention, the image processing method may be recorded on a computer-readable recording medium in which a program to be executed by a computer is recorded. As a result, it is possible to provide a portable recording medium on which a program for performing the image processing method is recorded.

  In the present embodiment, the recording medium may be a program medium such as a memory (not shown) such as a ROM itself because processing is performed by a microcomputer. However, it may be a program medium provided with a program reading device as an external storage device and readable by inserting a recording medium therein.

  In any case, the stored program may be configured to be accessed and executed by the microprocessor, or in any case, the program is read and the read program is illustrated in the microcomputer. The program may be downloaded to a non-program storage area and executed. It is assumed that this download program is stored in the main device in advance.

  Here, the program medium is a recording medium configured to be separable from the main body, and is a tape system such as a magnetic tape or a cassette tape, a magnetic disk such as a floppy (registered trademark) disk or a hard disk, or a CD-ROM / MO /. MD / DVD optical discs, IC cards (including memory cards) / optical cards, etc. Mask ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash A medium that carries a fixed program including a semiconductor memory such as a ROM may be used.

  In the present embodiment, the medium may be a medium that can be connected to a communication network including the Internet and fluidly carries the program so as to download the program from the communication network. When the program is downloaded from the communication network in this way, the download program may be stored in the main device in advance or may be installed from another recording medium.

  The above-mentioned image processing method is executed by reading the recording medium by a program reading device provided in a digital color copying machine or a computer system.

  The computer system includes an image input device such as a flatbed scanner, a film scanner, and a digital camera, a computer that performs various processes such as the above image processing method by loading a predetermined program, and a CRT display that displays the processing results of the computer. An image display device such as a liquid crystal display and a printer that outputs the processing results of the computer to paper or the like. Furthermore, a modem as a communication means for connecting to a server or the like via a network is provided.

  The present invention can be used in various devices such as a scanner device and a multi-function device that read a document once and perform image processing on the read document data.

1 is a block diagram illustrating a configuration of an input tone correction unit in an image processing apparatus of a color image forming apparatus according to an embodiment of the present invention. 1 is a cross-sectional view of a main part showing a schematic configuration of a color image forming apparatus according to the present invention. 1 is a block diagram showing a schematic configuration of a digital color copying machine to which an image processing apparatus of the present invention is applied. It is the graph which showed an example of the difference histogram. 3 is a flowchart illustrating an example of image processing in an input tone correction unit in FIG. 1. It is a block diagram which shows the structure of the input gradation correction | amendment part in the case of selecting the correction object area | region in the input gradation correction | amendment part in the said embodiment. FIG. 10 is a block diagram illustrating another embodiment of the present invention and illustrating a configuration of an input tone correction unit in a color image forming apparatus. It is a graph which shows an example of the density | concentration conversion table used with the input gradation correction | amendment part of FIG. It is a flowchart which shows the example of an image process in the input gradation correction | amendment part of FIG. (A) to (d) are respectively a graph showing a difference histogram, a graph showing a C histogram, a graph showing an M histogram, and a Y histogram created by the input tone correction unit of FIG. It is the graph which showed an example. (A) to (d) are respectively a graph showing a difference histogram, a graph showing a C histogram, a graph showing an M histogram, and a Y histogram created by the input tone correction unit of FIG. It is the graph which showed the other example. (A) to (d) are respectively a graph showing a difference histogram, a graph showing a C histogram, a graph showing an M histogram, and a Y histogram created by the input tone correction unit of FIG. It is the graph which showed other examples. 6 is a diagram illustrating an example of a document to be read. FIG. It is a flowchart which shows the other example of image processing in the input gradation correction | amendment part based on said other embodiment. It is a block diagram which shows the structure of the other input gradation correction | amendment part in a color image forming apparatus. 16 is a flowchart illustrating an example of image processing in another input tone correction unit in FIG. 15. It is a block diagram which shows the structure of the input gradation correction | amendment part in the case of selecting the correction | amendment object area | region in the input gradation correction | amendment part in the said other embodiment. FIG. 32 is a block diagram illustrating a configuration of an input tone correction unit in a color image forming apparatus according to still another embodiment of the present invention. 20 is a flowchart illustrating an example of image processing in an input tone correction unit in FIG. 19. (A), (b) is the graph which showed the difference histogram in the said further another embodiment, respectively, and the graph which showed an example of the brightness | luminance histogram. (A), (b) is the graph which showed the difference histogram in the said further another embodiment, respectively, and the graph which showed another example of the brightness | luminance histogram. (A), (b) is the graph which showed the difference histogram in the said further another embodiment, respectively, and the graph which showed another example of the brightness | luminance histogram. It is a flowchart which shows the other example of image processing based on the said further another embodiment. It is a block diagram which shows the structure of the other input gradation correction | amendment part in a color image forming apparatus. It is a flowchart which shows the example of an image process in the other input gradation correction | amendment part of FIG. (A), (b) is the graph which showed the difference histogram created by the input gradation correction | amendment part of FIG. 24, respectively, and the graph which showed an example of the brightness | luminance histogram. It is a block diagram which shows the structure of the input gradation correction | amendment part in the case of selecting the correction | amendment object area | region in the input gradation correction | amendment part in the said further another embodiment.

Explanation of symbols

1 Color image processing device (image processing device)
12 Region separation processing unit (region separation means)
13 Input gradation correction unit 30 Maximum value calculation unit (maximum value calculation means)
31 Minimum value calculation unit (minimum value calculation means)
32 Difference calculation unit (difference calculation means)
33 Difference histogram creation unit (difference histogram creation means)
34 Judgment part (background judgment means)
35 Conversion table selection unit 36 Density conversion unit 37 C histogram creation unit (color component histogram creation means)
38 M histogram creation unit (color component histogram creation means)
39 Y-histogram creation unit (color component histogram creation means)
40 CMY detection category calculation unit 41 Average value calculation unit 42, 42 'Shape determination unit (shape determination means)
43 Luminance value calculation unit (luminance value calculation means)
44 Luminance histogram creation unit (luminance histogram creation means)
45 Luminance detection category calculation unit 49/49 'Background level correction unit (background level correction means)
50 pixel selection unit (pixel selection means)

Claims (16)

In an image processing apparatus that determines at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
Maximum value calculating means for calculating the maximum value of the absolute value of the difference between the pixel values of each color component given to each pixel, for each pixel;
Maximum value histogram creating means for creating a histogram relating to the maximum value based on the maximum value;
Color component histogram creation means for creating a histogram relating to the pixel value of each color component based on the pixel value of each color component of each pixel;
The detection value is a frequency greater than a first threshold, which is first detected along the detection direction in the histogram relating to the maximum value, with the detection value from the smaller value to the larger value of the section. A first detection section that is a section having a frequency with which the sum of the frequencies of the sections adjacent to the second threshold is greater than the second threshold value is present in the section up to a predetermined upper limit value, and the pixel value of each color component In the histogram, the frequency that is first detected along the detection direction is greater than the third threshold, and the sum of the frequencies of the adjacent segments existing in the detection direction is greater than the fourth threshold. A background determination means for determining that there is a background when the value of the second detection section, which is a section having a frequency that is greater, is greater than the fifth threshold, and otherwise determining that there is no background;
When it is determined by the background determination means that there is a background, if the number of sections having a frequency greater than the frequency of the first detection section is smaller than a sixth threshold, When it is determined that the shape is sharp and the number of segments having a frequency greater than the frequency of the second detection segment is greater than or equal to a seventh threshold, Shape determining means for determining that the shape is broad;
When the shape determining means determines that the shape of the histogram related to the maximum value is sharp and the shape of the histogram related to the pixel value of each color component is broad, the shape determination unit determines whether the histogram is related to the value of the second detection section. An image processing apparatus comprising: a selection unit that selects a background removal amount larger than a background removal amount, and otherwise selects a background removal amount according to the value of the second detection category . In an image processing apparatus that determines at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
Among the pixel values of each color component given to each pixel, maximum value calculating means for calculating the maximum value for each pixel;
Among pixel values of each color component is applied to each pixel of the above, the minimum value calculating means for calculating a minimum value for each pixel,
For each of the above pixels , a difference calculation that calculates, for each pixel, the difference between the maximum value calculated by the maximum value calculation unit for the pixel and the minimum value calculated by the minimum value calculation unit for the pixel. Means,
A difference histogram creating means for creating a histogram relating to the difference based on the difference;
Color component histogram creation means for creating a histogram relating to the pixel value of each color component based on the pixel value of each color component of each pixel;
In the histogram relating to the difference, the value that is first detected along the detection direction is a frequency that is greater than the first threshold, and on the detection direction side. The first detection section, which is a section having a frequency with which the sum of the frequencies of the adjacent sections is greater than the second threshold, exists in the section up to a predetermined upper limit value, and relates to the pixel value of each color component The histogram has a frequency that is first detected along the detection direction and is greater than a third threshold value, and the sum of the frequencies of the adjacent segments on the detection direction side is greater than a fourth threshold value. A background determination unit that determines that the background is present when the value of the second detection classification that is the classification is larger than the fifth threshold ;
When it is determined by the background determination means that there is a background, when the number of sections having a frequency larger than the frequency of the first detection section is smaller than a sixth threshold, If the number of segments having a frequency greater than the frequency of the second detection segment is equal to or greater than a seventh threshold, the histogram shape relating to the pixel value of each color component is determined to be sharp. Shape determining means for determining that is broad,
The background corresponding to the value of the second detection section when the shape determining means determines that the shape of the histogram related to the difference is sharp and the shape of the histogram related to the pixel value of each color component is broad An image processing apparatus comprising: a selection unit that selects a background removal amount larger than the removal amount, and otherwise selects a background removal amount according to the value of the second detection section . In an image processing apparatus that determines at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
Maximum value calculating means for calculating the maximum value of the absolute value of the difference between the pixel values of each color component given to each pixel, for each pixel;
Minimum value calculation means for calculating the minimum value of the absolute value of the difference between the pixel values of each color component given to each pixel, for each pixel;
For each of the above pixels , a difference calculation that calculates, for each pixel, the difference between the maximum value calculated by the maximum value calculation unit for the pixel and the minimum value calculated by the minimum value calculation unit for the pixel. Means,
A difference histogram creating means for creating a histogram relating to the difference based on the difference;
Color component histogram creation means for creating a histogram relating to the pixel value of each color component based on the pixel value of each color component of each pixel;
In the histogram relating to the difference, the value that is first detected along the detection direction is a frequency that is greater than the first threshold, and on the detection direction side. The first detection section, which is a section having a frequency with which the sum of the frequencies of the adjacent sections is greater than the second threshold, exists in the section up to a predetermined upper limit value, and relates to the pixel value of each color component In the histogram, the frequency that is first detected along the detection direction is greater than the third threshold, and the sum of the frequencies of the adjacent segments existing in the detection direction is greater than the fourth threshold. A background determination means for determining that there is a background when the value of the second detection section, which is a section having a frequency, is greater than the fifth threshold, and otherwise determining that there is no background;
When it is determined by the background determination means that there is a background, when the number of sections having a frequency larger than the frequency of the first detection section is smaller than a sixth threshold, If the number of segments having a frequency greater than the frequency of the second detection segment is equal to or greater than a seventh threshold, the histogram shape relating to the pixel value of each color component is determined to be sharp. Shape determining means for determining that is broad,
The background corresponding to the value of the second detection section when the shape determining means determines that the shape of the histogram related to the difference is sharp and the shape of the histogram related to the pixel value of each color component is broad An image processing apparatus comprising: a selection unit that selects a background removal amount larger than the removal amount, and otherwise selects a background removal amount according to the value of the second detection section . In an image processing apparatus that determines at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
  Maximum value calculating means for calculating the maximum value of the absolute value of the difference between the pixel values of each color component given to each pixel, for each pixel;
  Maximum value histogram creating means for creating a histogram relating to the maximum value based on the maximum value;
  Luminance value calculating means for calculating the luminance value of each pixel;
  A luminance histogram creating means for creating a histogram relating to the luminance value based on the luminance value of each pixel;
  The detection value is a frequency greater than a first threshold, which is first detected along the detection direction in the histogram relating to the maximum value, with the detection value from the smaller value to the larger value of the section. In the histogram relating to the luminance value, there is a first detection section that is a section having a frequency whose sum with the frequency of the section adjacent to the second threshold is greater than the second threshold value. , A frequency that is first detected along the detection direction and that is greater than a third threshold value, and that is the sum of the frequencies of adjacent segments existing in the detection direction that is greater than the fourth threshold value. A background determination means for determining that there is a background when the value of the second detection section, which is a section having a value greater than the fifth threshold value, and determining that there is no background in other cases;
  When it is determined by the background determination means that there is a background, if the number of sections having a frequency greater than the frequency of the first detection section is smaller than a sixth threshold, When it is determined that the shape is sharp and the number of segments having a frequency greater than the frequency of the second detection segment is equal to or greater than a seventh threshold, the histogram shape is broad. Shape determining means for determining that
  When the shape determining unit determines that the histogram shape relating to the maximum value is sharp and the histogram shape relating to the luminance value is broad, the background removal amount according to the value of the second detection section An image processing apparatus comprising: a selecting unit that selects a larger background removal amount, and otherwise selects a background removal amount according to the value of the second detection section. In an image processing apparatus that determines at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
  Among the pixel values of each color component given to each pixel, maximum value calculating means for calculating the maximum value for each pixel;
  A minimum value calculating means for calculating a minimum value for each pixel among the pixel values of each color component given to each pixel;
  For each of the above pixels, a difference calculation that calculates, for each pixel, the difference between the maximum value calculated by the maximum value calculation unit for the pixel and the minimum value calculated by the minimum value calculation unit for the pixel. Means,
  A difference histogram creating means for creating a histogram relating to the difference based on the difference;
  Luminance value calculating means for calculating the luminance value of each pixel;
  A luminance histogram creating means for creating a histogram relating to the luminance value based on the luminance value of each pixel;
  In the histogram relating to the difference, the value that is first detected along the detection direction is a frequency that is greater than the first threshold, and on the detection direction side. In the histogram relating to the luminance value, the first detection section, which is a section having a frequency whose sum with the frequency of the adjacent section is larger than the second threshold, exists in the section up to a predetermined upper limit value. A frequency that is first detected along the detection direction and is greater than a third threshold value, and a frequency that is greater than a fourth threshold value with the frequency of adjacent segments existing in the detection direction. A background determination means for determining that there is a background when the value of the second detection section, which is a classification, is greater than the fifth threshold, and otherwise determining that there is no background;
  When it is determined by the background determination means that there is a background, when the number of sections having a frequency larger than the frequency of the first detection section is smaller than a sixth threshold, If the number of segments having a frequency greater than the frequency of the second detection segment is equal to or greater than a seventh threshold, the shape of the histogram is broad. Shape determining means for determining that there is,
  When the shape determining unit determines that the histogram shape relating to the difference is sharp and the histogram shape relating to the luminance value is broad, the background removal amount according to the value of the second detection section An image processing apparatus comprising: a selecting unit that selects a large amount of background removal, and otherwise selects a background removal amount according to the value of the second detection section. In an image processing apparatus that determines at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
  Maximum value calculating means for calculating the maximum value of the absolute value of the difference between the pixel values of each color component given to each pixel, for each pixel;
  Minimum value calculation means for calculating the minimum value of the absolute value of the difference between the pixel values of each color component given to each pixel, for each pixel;
  For each of the above pixels, a difference calculation that calculates, for each pixel, the difference between the maximum value calculated by the maximum value calculation unit for the pixel and the minimum value calculated by the minimum value calculation unit for the pixel. Means,
  A difference histogram creating means for creating a histogram relating to the difference based on the difference;
  Luminance value calculating means for calculating the luminance value of each pixel;
  A luminance histogram creating means for creating a histogram relating to the luminance value based on the luminance value of each pixel;
  In the histogram relating to the difference, the value that is first detected along the detection direction is a frequency that is greater than the first threshold, and on the detection direction side. In the histogram relating to the luminance value, the first detection section, which is a section having a frequency whose sum with the frequency of the adjacent section is larger than the second threshold, exists in the section up to a predetermined upper limit value. A frequency that is first detected along the detection direction and is greater than a third threshold value, and a frequency that is greater than a fourth threshold value with the frequency of adjacent segments existing in the detection direction. A background determination means for determining that there is a background when the value of the second detection section, which is a classification, is greater than the fifth threshold, and otherwise determining that there is no background;
  When it is determined by the background determination means that there is a background, when the number of sections having a frequency larger than the frequency of the first detection section is smaller than a sixth threshold, If the number of segments having a frequency greater than the frequency of the second detection segment is equal to or greater than a seventh threshold, the shape of the histogram is broad. Shape determining means for determining that there is,
  When the shape determining unit determines that the histogram shape relating to the difference is sharp and the histogram shape relating to the luminance value is broad, the background removal amount according to the value of the second detection section An image processing apparatus comprising: a selecting unit that selects a large amount of background removal, and otherwise selects a background removal amount according to the value of the second detection section. Area separating means for separating the read original into a plurality of areas including a character area and / or a halftone dot area;
A pixel selection unit that selects pixels of a predetermined region other than the character region and / or the dot region among the plurality of regions;
With respect to the pixel selected by the pixel selection unit, an image processing apparatus according to any one of claims 1 to 6, characterized in that the selection of the background removal amount. An image forming apparatus comprising the image processing apparatus according to any one of claims 1 to 7. In an image processing method for determining at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
A maximum value calculating step for calculating, for each pixel, a maximum absolute value of a difference between pixel values of each color component given to each pixel;
A maximum value histogram creating step of creating a histogram relating to the maximum value based on the maximum value;
A color component histogram creation step for creating a histogram regarding the pixel value of the color component for each color component based on the pixel value of each color component of each pixel;
The detection value is a frequency greater than a first threshold, which is first detected along the detection direction in the histogram relating to the maximum value, with the detection value from the smaller value to the larger value of the section. A first detection section that is a section having a frequency that is greater than the second threshold value and the sum of the frequencies of the sections adjacent to the second threshold is included in the sections up to a predetermined upper limit value, and relates to the pixel values of the respective color components In the histogram, the frequency that is first detected along the detection direction and that is greater than the third threshold value, and the sum of the frequencies of the adjacent segments on the detection direction side is greater than the fourth threshold value. A background determination step of determining that there is a background when the value of the second detection section that is a section having a value greater than the fifth threshold value, and otherwise determining that there is no background;
When it is determined that there is a background, when the number of segments having a frequency greater than the frequency of the first detection segment is smaller than the sixth threshold in the histogram regarding the maximum value, the shape of the histogram is sharp. When the number of sections having a frequency greater than the frequency of the second detection section is equal to or greater than a seventh threshold, the shape of the histogram is broad. A shape determination step for determining
When it is determined that the shape of the histogram relating to the maximum value is sharp and the shape of the histogram relating to the pixel value of each color component is broad, it is larger than the background removal amount corresponding to the value of the second detection section An image processing method comprising: selecting a background removal amount; otherwise, selecting a background removal amount according to the value of the second detection section . In an image processing method for determining at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
A maximum value calculating step for calculating a maximum value for each pixel among pixel values of each color component given to each pixel;
A minimum value calculating step for calculating a minimum value for each pixel among the pixel values of each color component given to each pixel;
For each pixel , the difference between the maximum value calculated in the maximum value calculation step for the pixel and the minimum value calculated in the minimum value calculation step for the pixel is calculated for each pixel. A difference calculating step to calculate;
A difference histogram creating step for creating a histogram relating to the difference based on the difference;
A color component histogram creation step for creating a histogram regarding the pixel value of the color component for each color component based on the pixel value of each color component of each pixel;
In the histogram relating to the difference, the value that is first detected along the detection direction is a frequency that is greater than the first threshold, and on the detection direction side. Histograms relating to pixel values of the respective color components, and the first detection segment, which is a segment having a frequency greater than the second threshold, in the sum of the frequencies of adjacent segments exists in the segment up to a predetermined upper limit value. A frequency that is first detected along the detection direction and is greater than a third threshold value, and the sum of the frequencies of the segments adjacent to the detection direction side is greater than a fourth threshold value. A background determination step for determining that there is a background when the value of the second detection section, which is a classification, is greater than the fifth threshold, and otherwise determining that there is no background;
When it is determined that there is a background, when the number of sections having a frequency greater than the frequency of the first detection section is smaller than the sixth threshold, the histogram relating to the difference is sharp. In addition, regarding the histogram relating to the pixel value of each color component, when the number of sections having a frequency greater than the frequency of the second detection section is equal to or greater than a seventh threshold, the shape of the histogram is broad. A shape determination step for determining;
When it is determined in the shape determination step that the shape of the histogram related to the difference is sharp and the shape of the histogram related to the pixel value of each color component is broad, the background removal according to the value of the second detection section A selection step of selecting a background removal amount larger than the amount, and selecting a background removal amount according to the value of the second detection section when it is determined in the shape determination step otherwise. Processing method. In an image processing method for determining at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
A maximum value calculating step for calculating, for each pixel, a maximum absolute value of a difference between pixel values of each color component given to each pixel;
A minimum value calculating step for calculating, for each pixel, a minimum absolute value of a difference between pixel values of each color component given to each pixel;
For each pixel , the difference between the maximum value calculated in the maximum value calculation step for the pixel and the minimum value calculated in the minimum value calculation step for the pixel is calculated for each pixel. A difference calculating step to calculate;
A difference histogram creating step for creating a histogram relating to the difference based on the difference;
A color component histogram creation step for creating a histogram regarding the pixel value of the color component for each color component based on the pixel value of each color component of each pixel;
In the histogram relating to the difference, the value that is first detected along the detection direction is a frequency that is greater than the first threshold, and on the detection direction side. Histograms relating to pixel values of the respective color components, and the first detection segment, which is a segment having a frequency greater than the second threshold, in the sum of the frequencies of adjacent segments exists in the segment up to a predetermined upper limit value. In any of the above, the frequency detected first along the detection direction is greater than the third threshold value, and the sum of the frequency of the adjacent segment on the detection direction side is greater than the fourth threshold value. A background determination step of determining that there is a background when the value of the second detection section, which is a section having a large frequency, is larger than the fifth threshold, and otherwise determining that there is no background;
When it is determined that there is a background, when the number of sections having a frequency greater than the frequency of the first detection section is smaller than the sixth threshold, the histogram relating to the difference is sharp. In addition, regarding the histogram relating to the pixel value of each color component, when the number of sections having a frequency greater than the frequency of the second detection section is equal to or greater than a seventh threshold, the shape of the histogram is broad. A shape determination step for determining;
When it is determined in the shape determination step that the shape of the histogram related to the difference is sharp and the shape of the histogram related to the pixel value of each color component is broad, the background removal according to the value of the second detection section A selection step of selecting a background removal amount larger than the amount, and selecting a background removal amount according to the value of the second detection section when it is determined in the shape determination step otherwise. Processing method. In an image processing method for determining at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
  A maximum value calculating step for calculating, for each pixel, a maximum absolute value of a difference between pixel values of each color component given to each pixel;
  A maximum value histogram creating step of creating a histogram relating to the maximum value based on the maximum value;
  A luminance value calculating step for calculating a luminance value of each pixel;
  A luminance histogram creating step of creating a histogram related to the luminance value based on the luminance value of each pixel;
  The detection value is a frequency greater than a first threshold, which is first detected along the detection direction in the histogram relating to the maximum value, with the detection value from the smaller value to the larger value of the section. In the histogram relating to the luminance value, a first detection segment that is a segment having a frequency greater than a second threshold is included in the segment up to a predetermined upper limit value. Section that is first detected along the detection direction and has a frequency that is greater than a third threshold value, and that has a frequency that is greater than the fourth threshold value with the sum of the frequencies of adjacent sections on the detection direction side. A background determination step for determining that there is a background when the value of the second detection classification is greater than the fifth threshold, and otherwise determining that there is no background;
  When it is determined that there is a background, when the number of segments having a frequency greater than the frequency of the first detection segment is smaller than the sixth threshold in the histogram regarding the maximum value, the shape of the histogram is sharp. When the number of segments having a frequency greater than the frequency of the second detection segment is equal to or greater than a seventh threshold, it is determined that the shape of the histogram is broad. A shape determination step;
  When it is determined that the shape of the histogram relating to the maximum value is sharp and the shape of the histogram relating to the luminance value is broad, the amount of background removal larger than the amount of background removal corresponding to the value of the second detection category And a selection step of selecting a background removal amount according to the value of the second detection section in the other cases. In an image processing method for determining at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
  A maximum value calculating step for calculating a maximum value for each pixel among pixel values of each color component given to each pixel;
  A minimum value calculating step for calculating a minimum value for each pixel among the pixel values of each color component given to each pixel;
  For each pixel, the difference between the maximum value calculated in the maximum value calculation step for the pixel and the minimum value calculated in the minimum value calculation step for the pixel is calculated for each pixel. A difference calculating step to calculate;
  A difference histogram creating step for creating a histogram relating to the difference based on the difference;
  A luminance value calculating step for calculating a luminance value of each pixel;
  A luminance histogram creating step of creating a histogram related to the luminance value based on the luminance value of each pixel;
  In the histogram relating to the difference, the value that is first detected along the detection direction is a frequency that is greater than the first threshold, and on the detection direction side. In the histogram relating to the luminance value, the first detection segment, which is a segment having a frequency that is greater than the second threshold, is included in the segment up to a predetermined upper limit value. A section that is first detected along the detection direction and has a frequency that is greater than a third threshold value, and that has a frequency that is greater than the fourth threshold value with the sum of the frequencies of the sections adjacent to the detection direction side. A background determination step of determining that there is a background if the value of a second detection category is greater than the fifth threshold; otherwise, determining that there is no background;
  When it is determined that there is a background, when the number of sections having a frequency greater than the frequency of the first detection section is smaller than the sixth threshold, the histogram relating to the difference is sharp. A shape for determining that the shape of the histogram is broad when the number of sections having a frequency greater than the frequency of the second detection section is greater than or equal to a seventh threshold in the histogram relating to the luminance value A determination step;
  When it is determined in the shape determination step that the shape of the histogram related to the difference is sharp and the shape of the histogram related to the luminance value is broad, it is larger than the background removal amount corresponding to the value of the second detection section An image processing method comprising: a selection step of selecting a background removal amount, and selecting a background removal amount according to the value of the second detection section when a background removal amount is selected and the shape determination step determines otherwise. In an image processing method for determining at least the presence or absence of a background of a document from image data of each pixel obtained by reading the document,
  A maximum value calculating step for calculating, for each pixel, a maximum absolute value of a difference between pixel values of each color component given to each pixel;
  A minimum value calculating step for calculating, for each pixel, a minimum absolute value of a difference between pixel values of each color component given to each pixel;
  For each pixel, the difference between the maximum value calculated in the maximum value calculation step for the pixel and the minimum value calculated in the minimum value calculation step for the pixel is calculated for each pixel. A difference calculating step to calculate;
  A difference histogram creating step for creating a histogram relating to the difference based on the difference;
  A luminance value calculating step for calculating a luminance value of each pixel;
  A luminance histogram creating step of creating a histogram related to the luminance value based on the luminance value of each pixel;
  In the histogram relating to the difference, the value that is first detected along the detection direction is a frequency that is greater than the first threshold, and on the detection direction side. In the histogram relating to the luminance value, the first detection segment, which is a segment having a frequency that is greater than the second threshold, is included in the segment up to a predetermined upper limit value. A section that is first detected along the detection direction and has a frequency that is greater than a third threshold value, and that has a frequency that is greater than the fourth threshold value with the sum of the frequencies of the sections adjacent to the detection direction side. A background determination step of determining that there is a background if the value of a second detection category is greater than the fifth threshold; otherwise, determining that there is no background;
  When it is determined that there is a background, when the number of sections having a frequency greater than the frequency of the first detection section is smaller than the sixth threshold, the histogram relating to the difference is sharp. A shape for determining that the shape of the histogram is broad when the number of sections having a frequency greater than the frequency of the second detection section is greater than or equal to a seventh threshold in the histogram relating to the luminance value A determination step;
  When it is determined in the shape determination step that the shape of the histogram related to the difference is sharp and the shape of the histogram related to the luminance value is broad, it is larger than the background removal amount corresponding to the value of the second detection section An image processing method comprising: a selection step of selecting a background removal amount, and selecting a background removal amount according to the value of the second detection section when a background removal amount is selected and the shape determination step determines otherwise. A program for causing a computer to function as each unit of the image processing apparatus according to any one of claims 1 to 7 . A computer-readable recording medium on which the program according to claim 15 is recorded.

Images