JP2009111786A - Image processing apparatus and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus for easily improving readability of an image including a low-density region such as a correction mark and storage costs of images, and to provide an image processing method. <P>SOLUTION: In an image processing apparatus, a background density value, a frequency of background density values, a low-density value and a frequency of low-density values are detected and a resolution of image data is determined in such a way that a resolution is reduced as the number of pixels within a predetermined range of density values around a correction mark density value becomes greater. Furthermore, compression processing is performed on the image data on the basis of the resolution determined. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to the field of image processing apparatuses and image processing methods.

  Generally, there is a background removal function as a function installed in an image processing apparatus. This is a function for improving the readability of an image when copying the document or viewing the image on a PC terminal or the like by removing the background of the document read by the scanner. On the other hand, if there is a portion corrected with a correction liquid or a correction tape in the document, the correction mark is also removed by the background removal processing. This is because the correction fluid or the like has a density lower than that of a document sheet that is generally used.

  By the way, in recent years, with the enforcement of the e-document law, various paper-based documents can be stored electronically. At this time, in the digitization or duplication of documents, it is necessary to ensure the sameness, authenticity, visibility and the like with the original document. Therefore, the image processing apparatus is also required to perform image processing that supports the correction marks held by the original document.

Patent Document 1 describes a background processing that leaves a low density region such as a correction mark. The density statistic of the image is taken, and shading correction is performed for an image including an object having a whiteness higher than the background to ensure that the image does not saturate, thereby correcting the brightness. Further, the correction mark is confirmed by enhancing the local spatial contrast on the brightness-corrected image.
JP 2006-203815 A

  The invention disclosed in Patent Document 1 provides image processing that is set by the user uniformly for all images regardless of whether a low density region such as a correction mark exists in the image or not. Is. For example, when multiple documents are digitized, it is considered that some of the documents do not have correction marks, but the same image processing that leaves correction marks even for images without such correction marks. Will be done. Unnecessarily leaving the background of the document in this way disturbs the regularity of the image data, reducing the compression efficiency and increasing the image data size. On the other hand, even if the presence / absence of processing is set for each image, when a large-scale document is processed, it is very laborious. Even if the lowest common multiple setting is made to reduce time and effort, it is not efficient.

  If processing for improving the readability of the correction trace and compression processing can be performed according to the feature of each image such as the presence or absence of the correction trace, the storage cost is improved as compared with the one that uniformly applies the same image processing.

  In view of the above problems, the present invention provides an image processing apparatus and an image processing method for performing image processing that easily improves the readability of an image including a low density region such as a correction mark and the storage cost of the image. For the purpose.

  Accordingly, in order to solve the above-described problem, an image processing apparatus according to the present invention includes an image processing apparatus having a reading unit that reads image data, and a density value having a maximum frequency of image density in the image data read by the reading unit. The background density value, the frequency of the background density value, the correction mark density value whose density value is lower than the background density value and the frequency is the second highest value after the frequency of the background density value, and the correction mark density value A density value detecting means for detecting the frequency, and the background density value detected by the density value detecting means, the correction mark density value, the frequency of the background density value, and the number of pixels based on the frequency of the correction mark density value. Resolution determining means for determining the resolution of the image data so that the number of pixels in a predetermined range of density values centered on the correction mark density value becomes lower as the density increases. Characterized in that it has.

  In order to solve the above problem, the image processing apparatus according to the present invention further includes compression processing means for performing compression processing of the image data based on the resolution determined by the resolution determination means. To do.

In order to solve the above problems, an image processing apparatus according to the present invention includes an image processing apparatus having a reading unit that reads image data, and a density value having a maximum frequency of image density in the image data read by the reading unit. A background density value detecting means for detecting the background density value,
On the basis of the background density value detected by the background density value detecting means, a correction mark density area for detecting the presence or absence of a correction mark density area which is a pixel area including a predetermined number of pixels having a density value lower than the background density value. The detection means, the distribution degree acquisition means for acquiring the distribution degree of the low density area in the image data, and the higher the resolution as the distribution degree of the correction mark density area acquired by the distribution degree acquisition means becomes higher. And a resolution determining means for determining the resolution of the image data.

  In order to solve the above-mentioned problem, in the image processing apparatus according to the present invention, the correction mark density area detection unit further divides the image data into predetermined areas and is lower than the background density value for each of the divided areas. The presence / absence of the correction mark density region is determined based on the number of pixels having the density value and the average density value.

  In order to solve the above-described problem, the image processing apparatus according to the present invention further includes the distribution degree acquisition unit that divides the image data into predetermined areas and detects the predetermined area detected as having the correction mark density area. The distribution degree of the correction mark density region is determined based on the number.

  In order to solve the above problem, the image processing apparatus according to the present invention further includes compression processing means for performing compression processing of the image data based on the resolution determined by the resolution determination means. To do.

  In order to solve the above-described problem, the image processing apparatus according to the present invention further includes: the image data based on a predetermined compression rate corresponding to the distribution degree of the correction mark density region acquired by the distribution degree acquisition unit. And compression processing means for performing compression processing.

  In addition, what applied the arbitrary combination of the component of this invention, expression, or a component to a method, an apparatus, a system, a computer program, a recording medium, etc. is also effective as an aspect of this invention.

  ADVANTAGE OF THE INVENTION According to this invention, the image processing apparatus and the image processing method which perform the image processing which improves the readability of the image containing the low density area | region like a correction trace and the preservation | save cost of an image can be provided easily.

  The best mode for carrying out the present invention will be described below with reference to the drawings in each embodiment.

(Outline and configuration of image processing apparatus)
First, the image processing apparatus according to the present embodiment will be described with reference to FIG. FIG. 1 shows an example of an overall configuration diagram of an image processing apparatus according to the present embodiment.

  The image processing apparatus 1 includes a reading device 2, a bus control device 3, a hard disk (HDD) 4, a CPU 5, a memory 6, a plotter I / F device 7, a plotter device 8, a display device 9, a line I / F device 10, an external I / F device 11, S.P. B. (South Bridge) 12 and ROM 13.

  The reading device 2 generates and outputs digital image data of 8 bits for each of RGB from the density information of the document obtained by scanning the document. The bus control device 3 exchanges various data such as image data and control commands necessary for the image processing device 1. The HDD 4 is a storage device for storing digital image data (including image data after image processing), incidental information of the digital image data, and the like. The CPU 5 is a microprocessor that controls the entire control of the image processing apparatus 1. The memory 6 is a volatile memory used for temporary storage of programs and intermediate processing data when the CPU 5 controls the image processing apparatus. The plotter I / F device 7 receives CMYK digital image data sent from the CPU 5 and outputs it to the plotter device 8. When the plotter device 8 receives the CMYK digital image data, the plotter device 8 outputs the image onto a transfer sheet by an electrophotographic process using a laser beam or the like. S. B. Reference numeral 12 denotes a general-purpose circuit that is a bridge function of a bus called South Bridge. The ROM 13 is a memory that stores a program when the CPU 5 controls the image processing apparatus 1 and processes image data. The display device 9 is a part that performs an interface between the image processing device 1 and the user, and is composed of an LCD (liquid crystal display device) and a key switch. The display device 9 displays various states and operation methods of the device on the LCD, and a key switch input from the user. Is detected. Further, the image processing apparatus 1 exchanges image data with a FAX 14 outside the apparatus via the line I / F apparatus 10 and a telephone line. Further, various controls and input / output of image data are performed via the PC terminal 15 outside the image processing apparatus 1, the external I / F apparatus 11, and a wired / wireless network.

(Example 1)
Next, Embodiment 1 of image processing performed by the image processing apparatus according to the present invention will be described with reference to FIG. FIG. 2 is a block diagram showing the image processing function of the image processing apparatus according to the present invention. The image data read by the reading device 2 includes a scanner γ conversion unit 21, a gradation correction unit 22, a filter processing unit 23, a background correction unit 24, a color correction unit 25, a density detection unit 26, a resolution determination unit 27, and a compression. After each image processing process in the processing unit 28, it is stored in the HDD 4 as RGB image data. Next, the processing of each unit will be described in the order of processing.

  The scanner γ conversion unit 21 performs γ conversion on RGB 8-bit image data read from the reading device 2 and converts the gradation characteristics from reflectance linearity to density linearity.

  The density detector 26 detects the background density, the correction mark density, and the frequency thereof from the image data. The background density is the density of the background of the original paper scanned and read. There are various types of manuscripts handled by scanners, fax machines, digital copiers, etc., and the density of the paper used for the manuscript, that is, the background density, is in a wide range. Further, the correction mark density is the density of the area of the correction portion that is painted white with a correction liquid or the like. In general, when correcting a typographical error or the like on a document sheet, the portion to be corrected is painted in white, or written in white and the like. This white-painted portion is a so-called correction mark, and the density of the correction mark is lower than the background density of the document. The density detection unit 26 acquires information on the density of the image and the frequency of the image data from the image data (a density histogram as shown in FIG. 5 can be obtained), and further calculates the background density, the correction mark density, and these frequencies. It is to detect.

  Here, in order to facilitate understanding of how the background density and the correction mark density are extracted when the density detection unit 26 acquires a density histogram corresponding to an image pattern to be described next from the image data, A density histogram corresponding to a pattern will be briefly described.

  FIG. 5 shows a so-called density histogram in which the image density and frequency of the RGB image data are detected. The horizontal axis represents image density and the vertical axis represents frequency. In the density histogram of FIG. 5, a protrusion (mountain) -like curve composed of a, b, and c is drawn. In the vicinity of the mountain b, the frequency of density is highest in the image, which indicates that there is a background density in the vicinity. a represents that there is a correction mark because it has a density lower than the background density and has a certain frequency. Since c has a density higher than the background density and has a constant frequency, c represents a character, a figure, or the like written on the document.

FIG. 6 is a density histogram representing an image without a correction mark. Since the density frequency is the highest in the vicinity of the mountain b, it represents the background density. Since c has a density higher than the background density and has a constant frequency, c represents a character, a figure, or the like written on the document. Compared to the case of FIG. 5, FIG. 6 shows a portion having a density lower than the background density and a frequency equal to or higher than a predetermined value (y), that is, a mountain corresponding to a correction mark (corresponding to a in FIG. 5). There is no. Therefore, the density histogram of FIG. 6 represents an image without a correction mark. In an actual image, there are a predetermined number of pixels having a density lower than the background density. In order to eliminate such errors and unevenness, y (predetermined frequency) is taken, and a density value that is lower than the background density and has a frequency of y or more is used as a correction mark. In FIG. 6, since the small mountain (a ₂ ) has a frequency of y or less, it is determined that there is no correction mark.

  FIG. 7 is formed from four peaks a, b1, b2, and c. 6 is a density histogram showing an extreme example in which the background of an original is uneven. Since b1 and b2 are uneven before and after the background density, two peaks with high frequency are formed. a represents a correction mark because it has a density lower than the background density and has a certain frequency. “c” represents characters on the original document as before.

  FIG. 8 shows a density histogram of an image having unevenness in the background because there is no correction mark because there is no mountain a which has a low frequency and a constant frequency, and has b1 and b2.

  Next, how the background density and the correction mark density are detected when the density detection unit 26 acquires these density histograms will be described.

In FIG. 5, the highest density in the histogram is the background density value Tr.
This is based on the fact that the widest area on the paper is the background (the background of the paper). Therefore, the density with the highest frequency, that is, the background density can be obtained. Further, a density having a density lower than the background density and the second highest frequency after the background density is set as a correction mark density value Tr. This is based on the fact that the density of the correction marks is lower than the background density of the document or paper, and has a certain frequency or more.

  In FIG. 6, the background density value is the density value Tr corresponding to the peak of the mountain b, but the correction mark density value is not detected because the image has no correction mark (no a).

  In the case where two peaks are formed near the background density due to uneven background as shown in FIG. 7, the density corresponding to the peak of the peak b2 corresponds to the background density value and the peak of the peak a as determined above. The density is determined as the correction mark density. However, the density corresponding to the peak of the peak b2 is actually the density in the vicinity of the background caused by unevenness of the background, and does not represent the density of the correction mark. For this reason, if the frequency is substantially the same even if it is determined as described above and once detected as the background density value and the correction mark density value, the low density value is determined as the background density (Tr). As a result, the density having a density lower than the density regarded as the background density value and the next highest frequency becomes the correction mark density (Tm) (the density corresponding to the peak of mountain a).

  In FIG. 8, the frequency of the detected background density value and the correction mark density value is the same as in FIG. 7, but even if a density value having a low density is regarded as the background density (Tr), it is next to the background density. There is no mountain (the frequency at the apex of a is y (predetermined frequency) or less). That is, since the image has no correction mark, the correction mark density value is not detected.

  Next, the gradation correction unit 22 performs gradation correction that increases the gradation of the portion of the correction mark in accordance with the density value information obtained from the density detection unit 26. The gradation correction method may be a known method. For example, when the correction mark density value is the background density value, no special correction is performed. When the correction mark density value is lower than the background density value, the scanner γ conversion unit The gradation correction is performed on the RGB image data generated from the image data 21 so as to increase the gradation of the highlight portion.

  The filter processing unit 23 is a process performed on the image data obtained from the gradation correction unit 22. Filter processing such as emphasizing the character part or smoothing the picture part is performed.

  The background correction unit 24 performs background correction on the RGB image data obtained from the filter processing unit 23. The background correction is to remove unnecessary unevenness below the background density of a document sheet having various densities. In the case of an image without a correction mark, by correcting the background, it is possible to eliminate irregularities in the image data and make the image data regular, which contributes to an improvement in compression processing and storage cost. On the other hand, in the case of an image having a correction mark, if the background correction is performed, the correction mark is erased or difficult to see, and the correction mark cannot be left appropriately. Therefore, the background correction is not performed or the background correction is performed with a correction mark left. For example, the unevenness of the background having a density lower than the correction mark is corrected.

  The color correction unit 25 is a process performed on the image data obtained from the background correction unit 24. Conversion into RGB data of a color space having a predetermined characteristic. This color space is, for example, an Adobe-RGB space defined by Adobe. This may be performed by a known general technique.

  The resolution determination unit 27 is a process performed on the image data obtained from the color correction unit 25. The resolution determination unit 27 determines the resolution of the image data based on the background density value (Tr) and the correction mark density value (Tm) obtained by the density detection unit 26. Next, the resolution determination method will be specifically described.

The resolution determination unit 27 is a histogram of density acquired by the density detection unit 26 (for example, FIG. 5),
The background density value (Tr) correction mark density value (Tm) value is acquired. Then, the frequency f (x) at the image density x is integrated over the interval [I, J]

としたとき、S_{Tm-α,Tm+α}（αは一定の値で0<α<<Tm）は、修正痕濃度付近の画素数がどの程度あるかを数値化したものになる。そして、S_{Tm-α,Tm+α}の値が大きいということは、修正痕濃度の画素数が多いことを意味する。即ち、これは原稿に修正痕のある部分が多く修正箇所がユーザが見ても明らかである場合を意味している。よって、この場合は、解像度が低くても修正痕の視認が容易であるため解像度を下げる、または現状維持すればよいことになる。この解像度を下げるというのは画像の精細さを下げる分、データ容量が減るので画質と画像データ削減とのトレードオフの関係となる。

S_ {Tm−α, Tm + α} (α is a constant value and 0 <α << Tm) is a numerical value indicating how many pixels are near the correction mark density. A large value of S_ {Tm-α, Tm + α} means that the number of pixels of the correction mark density is large. That is, this means that there are many portions with correction marks on the original and the correction portions are clear even when viewed by the user. Therefore, in this case, even if the resolution is low, it is easy to visually recognize the correction mark, so the resolution may be lowered or the current state may be maintained. Lowering the resolution has a trade-off relationship between image quality and image data reduction because the data capacity is reduced as the image definition is lowered.

  On the other hand, when the value of S_ {Tm-α, Tm + α} is small, the number of pixels of the correction mark density is small. That is, it means that there are few correction marks and it is difficult to visually recognize the correction marks at the current resolution. Therefore, in this case, the resolution may be increased so that the correction marks can be seen well.

  For example, if the value of S_ {Tm-α, Tm + α} is x, the resolution corresponding to the x value is determined as shown in FIG. As the x value increases, the resolution decreases. To illustrate using realistic values, if the original image is equivalent to 600 dpi and S_ {Tm-α, Tm + α} = 60, the resolution is in the range of 400 dpi from FIG. It is decided to have a resolution of 400 dpi to enter.

  Thereby, according to the density information of the image, it is possible to automatically determine the resolution so that the correction mark density portion can be confirmed more easily than the background density such as the correction mark. Moreover, the readability of the image containing the correction trace whose density is lower than the background can be improved.

  The compression processing unit 28 compresses the RGB data according to the resolution determined by the resolution determination unit 27. In the compression process, compression is performed by a known method and accumulated in the HDD 4 as RGB image data. It should be noted that an appropriate compression rate corresponding to the resolution is determined empirically as the compression rate applied when compressing image data. Therefore, when a certain resolution is determined based on this, the compression may be performed using a compression rate corresponding to the resolution. The resolution obtained by the resolution determination unit 27 is determined as a resolution at which correction marks can be easily confirmed. Compared to the case where compression processing is performed using a compression rate corresponding to the resolution (for example, using a uniform resolution) that is determined regardless of the presence or absence of the correction mark, a compression rate suitable for the resolution at which the correction mark can be easily confirmed. The amount of image data can be reduced if compression processing is performed using. That is, the image storage cost can be improved by changing the compression rate to an appropriate compression rate according to the resolution obtained in the resolution determination unit 27.

  With the above configuration, the user can obtain an image with an appropriate resolution without the need for complicated settings when digitizing paper text. Further, the storage cost can be improved when storing the image data.

(Example 2)
Next, a second embodiment of image processing performed by the image processing apparatus according to the present invention will be described with reference to FIG. FIG. 3 is a block diagram showing the image processing functions of the image processing apparatus according to the present invention in the order of processes. The image data read by the reading device 2 includes a scanner γ conversion unit 21, a correction mark density region detection unit 32, a background density detection unit 33, a gradation correction unit 34, a filter processing unit 23, a background correction unit 36, and a color correction unit. 25, each image processing process in the resolution determination unit 38 and the compression processing unit 39 is stored in the HDD 4 as RGB image data. Next, the processing of each unit will be described in the order of processing. Note that the scanner γ conversion unit 21, the filter processing unit 23, and the color correction unit 25 are the same as already described in the first embodiment, and thus description thereof is omitted.

  The background density detector 33 detects the background density of the RGB image data generated from the scanner γ converter 21. As described with reference to FIG. 5 of the first embodiment, the image density histogram is taken, and the image density value Tr having the peak frequency is set as the background density.

  The correction mark density area detection unit 32 detects the presence or absence of the correction mark density area based on the background density obtained by the background density detection part 33. First, the correction mark density region detection unit 32 divides image data into blocks and searches for pixels locally. A known method may be used as the search method. For example, the image of the RGB image data generated from the highly accurate scanner γ converting unit 21 is divided into arbitrary blocks, and the histogram for each block is compared with the background density obtained from the background density detecting unit 33 to obtain the background. Also, the presence / absence of a correction mark density region is determined. When the background density value of the document is equal to the density value of the correction mark density area, it is determined that there is no correction mark density area of the image because the density becomes the background density. Therefore, the background is left and there is no loss of images.

  Here, a specific determination method for the presence / absence of the correction mark density region is illustrated. First, the image data to be processed is divided into blocks. Here, each divided block includes n × m pixels (n and m are arbitrary integers) as shown in FIG. At this time, the density value of each pixel in the block is stored in a memory P (p_ {n, m}) of n rows and m columns prepared in advance. Specifically, for example, the density value of a pixel in k rows and l columns (k: 1 to n integer, l: 1 to m integer) in the block is set as an element p_ {k, l} in the memory P. . This is repeated for all the pixels in the block. Further, a memory Q having the same size as the memory P is prepared, and the result of comparing the density value of each pixel stored in the memory P and the background density value Tr is stored in the memory Q (q_ {n, m}). Specifically, for example, considering the element p_ {k, l} of the memory P, if Tr <p_ {k, l} (when the density is higher than the background density), q_ {k, l} = 0, Tr > = p_ {k, l} (when the density is lower than the background density), q_ {k, l} = 1 ([k: 1,2 ... n, l: 1,2, ...・, M]). FIG. 11 shows the memory Q. Each element of the memory Q has a value of 0 meaning that the density is higher than the background density, or 1 meaning that the density is lower than the background density. This is repeated for all the pixels in the block.

  From this, the frequency of pixels with a lower density (corresponding to the correction mark density) than the background density (Tr) in the block is

となる（地肌濃度より低い濃度を持つ画素の総和）。また、ブロック内の低濃度の画素全体の平均的な濃度値は、

(The sum of pixels having a density lower than the background density). Also, the average density value of all low density pixels in the block is

となる。このようにして、処理対象の画像データをブロック単位で分割したブロック内の、地肌濃度より低濃度濃度の画素の頻度と平均濃度値を求めることができる。

It becomes. In this manner, the frequency and average density value of pixels having a density lower than the background density in the block obtained by dividing the image data to be processed in block units can be obtained.

  Subsequently, the statistic of the entire image data to be processed (that is, all blocks) is obtained. The presence / absence of a correction mark density region is determined from the value. Here, when the number of blocks of image data is z, and the frequency and the average density value of arbitrary blocks are H_b, I_b [b: 1, 2,..., Z], respectively, Statistics,

と定義する。これは、画像データに含まれる各ブロックにおいて前記平均濃度値と地肌濃度値の差分（Tr-I_b）に前記頻度（H_b）を乗算したものの総和である。これは、地肌濃度付近の濃度は読取り装置２から読取られたＲＧＢ画像データの測定誤差部分も含まれる可能性があることを想定し、平均濃度値と地肌濃度値の差分をとることにより低濃度部分に対して重み付けを行うことを考慮しての統計量である。

It is defined as This is the total sum of the difference (Tr-I_b) between the average density value and the background density value multiplied by the frequency (H_b) in each block included in the image data. It is assumed that the density near the background density may include a measurement error portion of the RGB image data read from the reading device 2, and the low density is obtained by taking the difference between the average density value and the background density value. This is a statistic taking into account the weighting of the part.

が所定値以上であれば（例えば、

Is greater than or equal to a predetermined value (for example,

εαは読取装置の性能等や環境に応じて決定した値）、低濃度領域（修正痕濃度領域）が有ると判断し、満たさない場合、低濃度領域（修正痕濃度領域）が無いと判断する。

εα is a value determined according to the performance of the reading device and the environment and the environment), and it is determined that there is a low density area (corrected trace density area). If not, it is determined that there is no low density area (corrected trace density area). .

  The gradation correction unit 34 performs gradation correction on the RGB image data generated from the scanner γ conversion unit 31 according to the presence or absence of the correction mark density region of the image determined by the correction mark density region detection unit 33. In the correction method, for example, no special correction is performed when there is no correction mark density region, and gray level correction is performed so as to increase the gradation of the highlight portion of the RGB image data when there is a correction mark density region.

  The background correction unit 36 is notified of the presence or absence of the correction mark density region of the image from the correction mark density region detection unit 32, and obtains the background density value (Tr) from the background density detection unit 33. If there is no correction mark density region, background correction may be performed by a known technique or the like. In the case of an image without a correction mark, by correcting the background, it is possible to eliminate irregularities in the image data and make the image data regular, which contributes to an improvement in compression processing and storage cost. On the other hand, if there is a correction mark density region, the background correction is not performed or the background correction is performed with the correction mark left. For example, the unevenness of the background having a lower density than the correction mark (corresponding to the Tm value) is corrected.

  Next, the resolution determination unit 38 will be described. The correction mark density region detection unit 32 gives the resolution determination unit 38 information on the presence / absence of the correction mark density region and, if there is a correction mark density region, information on the block having the correction mark density region in each block. If there is no correction mark density region, the resolution determination unit 38 may use the resolution specified by the user on the display device 9.

  When there is a low density area (corrected trace density area), if the number of blocks having the low density area among z blocks is z ′, the ratio of the low density area in the block in the block X_num (num: 1 ~ Z 'integer)

とおける。

You can.

  Next, the ratio of the low density area to the entire block with the low density area (here, z 'blocks)

とすると、

Then,

は低濃度領域があるブロック全体における低濃度領域（修正痕濃度領域）のばらつき度合いを見ることができる値（分布度）となる。

Is a value (distribution degree) that allows the degree of variation of the low density area (corrected trace density area) in the entire block having the low density area to be seen.

  Here, when the value of σ is large, there may be a case where there is a block having a large | X_num-μ |. This means that low density regions for a specific block are concentrated, and there is a high possibility that a small correction mark is present at a specific location.

  When the value of σ is small, there may be a case where there is no block with a large | X_num-μ |. It can be expected that the correction marks themselves are large because the correction marks are distributed as a whole.

  Thus, according to the value of σ, it is possible to determine the degree of variation of the low density region in the entire block having the low density region.

  Therefore, according to this σ value, the resolution may be increased if the σ value is large, and the resolution may not be changed if the σ value is small. If the σ value is small, processing for increasing the resolution is performed to make it easier to see the correction mark.

  The calculation process of the σ value may be performed by the correction mark density region detection unit 32. The resolution determination unit 38 may determine the resolution based on the σ value obtained from the correction mark density region detection unit 32.

  Thereby, according to the information of the correction mark density region from the background of the image, it is possible to automatically perform an appropriate resolution determination process for a portion having a lower density than the background such as the correction mark. In addition, this makes it possible to improve the readability of an image including a correction mark. The processing in the first embodiment has an advantage that the number of circuits (circuits related to correction mark density region detection) can be reduced, whereas the present embodiment has an advantage that the accuracy of resolution determination processing is increased. is there.

  Since the compression processing unit 39 obtains the resolution determined by the resolution determination unit 38 and changes the image data to an appropriate compression rate, the image storage cost can be improved. The image data is compressed and stored in the HDD 4 as RGB image data. The compression method may be the same as that of the compression processing unit 28.

  With the above configuration, the user can obtain an image with an appropriate resolution without requiring complicated settings when digitizing paper text. Further, the storage cost can be improved when storing the image data.

(Example 3)
Next, a third embodiment of image processing performed by the image processing apparatus according to the present invention will be described with reference to FIG. FIG. 4 differs from FIG. 3 of the second embodiment only in the processing content in the compression processing unit c40. Since the other parts (scanner γ conversion unit 21 to resolution determination unit 38) are the same as the contents described in the second embodiment, description thereof is omitted.

  The compression processing unit c40 compresses the RGB image data and accumulates the RGB image data in the HDD 4 as RGB image data. The compression processing unit c40 determines the compression rate based on the σ value determined by the resolution determination unit 38 and performs compression.

FIG. 12 shows an appropriate compression rate corresponding to the σ value. This is a compression rate in consideration of the degree of data variation in image data. What is necessary is just to compress using the compression rate according to (sigma) value.
As described above, by performing compression based on the variation degree of the correction marks, it is possible to obtain an image size more suitable for an image having the correction marks. For example, a compression ratio of 90 is applied if σ <B1, and 80 if B1 <σ = <B2. QF represents the compression ratio in the case of JPEG, and the higher the value, the lower the compression ratio.

  According to the present embodiment, as compared with the compression processing processed in the second embodiment, the compression rate is changed to an appropriate compression rate according to the changed resolution and the information of the region having a lower density than the background. As a result, the accuracy of the compression process is increased and the image storage cost can be improved.

  With the above configuration, the user can obtain a compressed image using an appropriate resolution without requiring complicated settings when digitizing paper text. Further, the storage cost can be improved when storing the image data.

  As mentioned above, although the Example of this invention was explained in full detail, this invention is not limited to the specific embodiment which concerns, In the range of the summary of this invention described in the claim, various deformation | transformation * It can be changed.

It is a hardware block diagram of the image processing apparatus which concerns on this embodiment. It is an example of the block diagram which concerns on an image process. It is an example of the block diagram which concerns on an image process. It is an example of the block diagram which concerns on an image process. It is an example of the figure showing an image density histogram. It is an example of the figure showing an image density histogram. It is an example of the figure showing an image density histogram. It is an example of the figure showing an image density histogram. It is an example of the figure showing the resolution corresponding to x value. It is an example showing the figure of a block. It is an example of the figure showing the pixel which has a low density value in a block. It is an example showing the correspondence between the σ value and the compression rate.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 26 Density detection part 27 Resolution determination part 28 Compression processing part 32 Low density area | region detection part 33 Background density detection part 40 Compression processing part

Claims (10)

In an image processing apparatus having a reading unit for reading image data,
In the image data read by the reading unit, the background density value having the highest image density frequency, the frequency of the background density value, and the frequency lower than the background density value is the frequency of the background density value. A correction mark density value that is the next highest density value, and a density value detection means for detecting the frequency of the correction mark density value;
It is the number of pixels based on the background density value detected by the density value detection means, the correction mark density value, the frequency of the background density value, and the frequency of the correction mark density value, and the correction mark density value is centered. Resolution determination means for determining the resolution of the image data so that the number of pixels in a predetermined range of density values becomes lower as the number of pixels increases.
An image processing apparatus comprising: When the frequency of the background density value and the frequency of the correction mark density value are substantially the same, the density value detection means sets the density value having the lower density as the background density value,
The image processing apparatus according to claim 1. Compression processing means for performing compression processing of the image data based on the resolution determined by the resolution determination means;
The image processing apparatus according to claim 1, further comprising: In an image processing apparatus having a reading unit for reading image data,
A background density value detecting means for detecting a background density value which is a density value having the highest frequency of image density in the image data read by the reading unit;
On the basis of the background density value detected by the background density value detecting means, a correction mark density area for detecting the presence or absence of a correction mark density area which is a pixel area including a predetermined number of pixels having a density value lower than the background density value. Detection means;
A distribution degree acquisition means for acquiring a distribution degree of the low density region in the image data;
Resolution determination means for determining the resolution of the image data so that the resolution becomes higher as the distribution degree of the correction mark density region acquired by the distribution degree acquisition means increases;
An image processing apparatus comprising: The correction mark density area detection unit divides the image data into predetermined areas, and the correction mark density is based on the number of pixels having a density value lower than the background density value and an average density value for each of the divided areas. Determining the presence or absence of an area,
The image processing apparatus according to claim 4. The distribution degree obtaining means divides the image data into predetermined areas, and determines the distribution degree of the correction mark density area based on the number of the predetermined areas detected as having the correction mark density area;
The image processing apparatus according to claim 4. Compression processing means for performing compression processing of the image data based on the resolution determined by the resolution determination means;
7. The image processing apparatus according to claim 4, further comprising: Compression processing means for compressing the image data based on a predetermined compression rate corresponding to the distribution degree of the correction mark density region acquired by the distribution degree acquisition means;
7. The image processing apparatus according to claim 4, further comprising: In an image processing method in an image processing apparatus having a reading unit for reading image data,
In the image data read by the reading unit, the background density value having the highest image density frequency, the frequency of the background density value, and the frequency lower than the background density value is the frequency of the background density value. A correction mark density value that is the next highest density value, and a density value detection procedure for detecting the frequency of the correction mark density value;
The background density value detected by the density value detection procedure, the correction mark density value, the frequency of the background density value, and the number of pixels based on the frequency of the correction mark density value, the correction mark density value being the center A resolution determination procedure for determining the resolution of the image data so that the number of pixels in a predetermined range of density values becomes lower as the number of pixels increases.
An image processing method comprising: In an image processing method in an image processing apparatus having a reading unit for reading image data,
A background density value detection procedure for detecting a background density value that is the density value having the highest frequency of image density in the image data read by the reading unit;
Based on the background density value detected by the background density value detection procedure, a correction mark density area that detects the presence or absence of a correction mark density area that is a pixel area that includes a predetermined number of pixels having a density value lower than the background density value. Detection procedure;
A distribution degree acquisition procedure for acquiring a distribution degree of the low density region in the image data;
A resolution determination procedure for determining the resolution of the image data so that the resolution becomes higher as the distribution degree of the correction mark density region acquired by the distribution degree acquisition procedure increases;
An image processing method comprising:

Images