JP3393433B2 - Image processing apparatus and method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus and method such as an image scanner.

[0002]

2. Description of the Related Art Conventionally, in an image processing apparatus for determining black and white for each pixel forming an image, binary processing is performed to process a character image, and a photographic (dot image) image is processed. Has undergone halftone processing such as dither processing. Usually, these processes are switched manually.

It has also been proposed to automatically distinguish between a character image and a photographic image, but it is configured to clearly distinguish whether binary processing or halftone processing is to be performed and to switch processing methods. ing.

[0004]

However, the actual image includes various images such as photographs with high contrast, large characters, small characters, etc., and binary processing and halftone processing are required as in the conventional art. There is a problem in that the image quality is deteriorated if the distinction is made and switching is performed.

The present invention has been made in view of the above circumstances, and an object thereof is to provide an image processing apparatus and method capable of improving the image quality.

[0006]

First image processing of the present invention
The device divides the image into blocks containing a predetermined number of pixels
Method and halftone processing and binary processing according to the type of image
For halftone processing among multiple image processing methods including
A fitness calculation means for calculating the fitness of a block,
Select a processing method according to the degree of compatibility from the image processing methods
And a processing method selecting means for performing the processing method selecting means.
Select halftone processing when the goodness of fit is the maximum value, and the goodness of fit is
When the minimum value is selected, binary processing is selected, and the goodness of fit is the maximum value.
When taking values between the minimum values, halftone processing and binary processing
It is characterized in that and are selected at a frequency according to the goodness of fit.

The second image processing apparatus of the present invention stores an image.
A method of dividing into blocks containing a constant number of pixels and the seed of the image
Multiple images including halftone processing and binary processing according to class
The degree of suitability of the block for binary processing among the processing methods
Among the plurality of image processing methods, the fitness calculation means for calculating
Select the processing method according to the degree of conformance from
The processing method selecting means is provided with
Select binary processing, and when the goodness of fit is the minimum value, it is intermediate
Select the adjustment process, and set the goodness of fit to a value between the maximum and minimum values.
The binary processing and the halftone processing according to the degree of conformity.
It is characterized by selecting by frequency.

A third image processing apparatus of the present invention stores an image.
A method of dividing into blocks including a constant number of pixels, and binary processing
And the block of halftone processing for halftone processing
The compatibility calculation means for calculating the compatibility and the compatibility calculation means
Threshold calculation to calculate the threshold according to the calculated fitness
And the degree of conformity calculated by the threshold calculation means.
Use the threshold and halftone processing threshold to
And a threshold value determining means for determining a threshold value for
To collect.

A fourth image processing apparatus according to the present invention is provided with a plurality of images.
Divide a plain image into blocks containing a certain number of pixels
Of the text area and background area
A fitness calculation means for calculating the fitness of the corresponding block,
Either fixed binary processing or floating binary processing depending on the goodness of fit
It is equipped with a processing method selection means for selecting one of them
The selection means selects floating binary processing when the goodness of fit is maximum.
If the goodness of fit is the minimum value, select fixed binary processing,
Float when the goodness of fit takes a value between the maximum and minimum values.
Binary processing or fixed binary processing is selected at a frequency according to the goodness of fit.
It is characterized by doing.

A fifth image processing device of the present invention is a device for processing a plurality of images.
Divide an image containing primes into blocks containing a certain number of pixels
Method and calculating the degree of conformity of the block to the character area
Fitness calculation means, fixed binary threshold and floating binary processing
The weighted average of the threshold value and the
Threshold value deciding unit that uses the average value of
And a step.

The first image processing method according to the present invention is a method for processing an image.
A step of dividing into blocks containing a constant number of pixels,
Of binary processing or halftone processing, for halftone processing
A fitness calculation step to calculate the fitness of the block, and
Depending on the goodness of fit calculated by the processing of the goodness of fit calculation step
Threshold calculation step for calculating the
Threshold and medium according to the goodness of fit calculated by the
Use the threshold for intertone processing and the threshold for black and white determination.
And a step of determining a threshold value to be determined.
It

A second image processing method of the present invention is a method for processing a plurality of images.
Divide an image containing primes into blocks containing a certain number of pixels
The division step and the conformity of the block to the character area
Goodness of fit calculation step to calculate, and threshold of fixed binary processing
Floating binary processing threshold and weighted average according to goodness of fit
Then, this weighted average value is used as a threshold for the pixel to be processed.
And a step of determining a threshold value.

[0013]

[0014]

[0015]

[0016]

[0017]

[0018]

[0019]

[0020]

[0021]

[0022]

[0023]

[0024]

[0025]

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

In the first image processing apparatus of the present invention, the image
Is divided into blocks containing a certain number of pixels, and halftone processing is performed.
The goodness of fit of the block is calculated, and the goodness of fit is the maximum.
When halftone processing is selected and the goodness of fit is the minimum value
Is selected and the goodness of fit is between the maximum and minimum values.
When taking the value of, the halftone processing and the binary processing
Since it is selected at a frequency according to
Which processing is suitable for the pixel data of
You can improve the quality of images that cannot be judged evenly.
It In addition, even if the fitness changes, all pixels that make up the image
Since it does not affect the
Compared with the method that affects the threshold, the concentration changes slowly.
The image quality does not change suddenly in the changing part.

In the second image processing apparatus of the present invention,
The image is divided into blocks containing a certain number of pixels
The block's goodness of fit is calculated and the goodness of fit is maximized.
When binary processing is selected for the value and the goodness of fit is the minimum value
Halftone processing is selected for the
Binary processing and halftone processing are compatible when taking values between
Since it is selected at a frequency according to the degree, block
Which processing is suitable for only the pixel data in
You can improve the quality of images that cannot be determined uniformly.
It In addition, even if the fitness changes, all pixels that make up the image
Since it does not affect the
Compared with the method that affects the threshold, the concentration changes slowly.
The image quality does not change suddenly in the changing part.

A third image processing apparatus and a first image processing apparatus according to the present invention.
In the image processing method, the image contains a certain number of pixels.
Divided into locks, blocks adapted for halftone processing
Is calculated, and a threshold value is calculated according to the calculated fitness.
The threshold and halftone processing according to the calculated goodness of fit.
Threshold and is used to determine the threshold for black and white decision
As a result, halftone processing is applied to pixels with intermediate brightness.
The binary processing is applied to high-luminance and low-luminance pixels.
To get a sharp-edged image
be able to.

In the fourth image processing apparatus of the present invention,
A block in which an image composed of multiple pixels contains a predetermined number of pixels
Is divided into, and the conformity of the block to the character area is calculated
Is selected and floating binary processing is selected when the goodness of fit is the maximum value.
If the goodness of fit is the minimum, fixed binary processing is selected.
And the goodness of fit takes a value between the maximum and minimum values,
Floating binary processing and fixed binary processing at a frequency according to the degree of compatibility
Since it is selected, it is the pixel data in the block
In this case, it is impossible to uniformly determine which process is suitable.
It is possible to improve the image quality of a bad image. Also, the goodness of fit
Even if the value changes, it may affect all the pixels that make up the image.
Since there is no, the change in fitness affects the thresholds of all pixels.
Image quality in areas where the density changes gently compared to
Does not change suddenly.

A fifth image processing apparatus and a second image processing apparatus according to the present invention.
In the image processing method, an image including a plurality of pixels is predetermined.
Divided into blocks containing several pixels,
The block goodness of fit is calculated, and the fixed binary processing threshold and floating
The dynamic binary processing threshold is a weighted average depending on the goodness of fit.
This weighted average value becomes the threshold value for the pixel to be processed.
Fixed binary processing depending on the goodness of fit.
And processing that has intermediate characteristics of floating binary processing
Since it can be,
Image quality that cannot be uniformly determined whether
Can be improved.

[0039]

[0040]

[0041]

[0042]

[0043]

[0044]

[0045]

[0046]

[0047]

[0048]

[0049]

[0050]

FIG. 1 shows the configuration of a first embodiment of an image processing apparatus according to the present invention. The image data capturing device 1 supplies the image count to the dither number calculating device 2 and also supplies the image data to the block unit dither compatibility calculating device (compatibility calculating means) 3, the binary processing device 4 and the dither processing device 5. To do. The dither number calculation device 2 calculates the dither number from the image count and supplies it to the binary dither selection flag calculation device (processing method selection means) 6. Binary processing device 4
Outputs the binary processing result for the image data.
Supply to. The dither processing device 5 supplies the dither processing result for the image data to the output determining device 7.

The block unit dither compatibility calculation device 3 is
All pixels forming the image are divided into blocks of 8 pixels x 1 column as shown in Fig. 3, and the maximum edge and the number of pulses are calculated from the data of 8 pixels in the block to be processed or the data of 32 pixels including the surrounding blocks. The feature amount of the block to be processed is calculated, and fuzzy inference is performed using the feature amount as an input to obtain the degree of suitability of the block to be processed for dither processing. Here, the goodness of fit is represented by five values as follows. Goodness of fit = {0.00, 0.25, 0.50, 0.75, 1.00} However, in actual hardware, goodness of fit = {0, 1,
2, 3, 4} may be used.

The binary dither selection flag calculating device 6 receives the dither number from the dither number calculating device 2 and the dither matching degree from the block unit dither matching degree calculating device 3, and outputs a selection flag to the output determining device 7. Then, the binary processing result and the dither processing result are selected at a frequency according to the following degree of conformity. Goodness of fit 0 All Binary processing Goodness of fit 1 4 times Dithering suitability 2 2 times 4 4 Dithering suitability 3 4 3 times Dithering suitability 4 All dithering

FIG. 4 shows the currently used dither matrix. As is clear from FIG. 4, since the dither matrix has a period of 4, if the remainder is processed as it is, the area to be binarized and the area to be dithered will be separated. Therefore, it is preferable to change the cycle. FIG. 2 shows an example of the binary dither selection flag calculation device 6 configured in consideration of this point. Register 61
Holds a 6-bit dither number N. The register 62 is
A 3-bit comparator 6 which holds the 3-bit dither compatibility G
Output to 7. The register 63 receives the output of the register 61 and holds the middle 2 bits B of the dither number N. The register 64 receives the output of the register 61 and holds the lower 2 bits A of the dither number N. 2-bit adder 65
Adds the contents of registers 63 and 64. However, it is not carried forward. The register 66 outputs the output of the adder 65,
That is, the lower 2 bits C of (A + B) are held. The register 66 always holds 0 as the most significant bit. The comparator 67 compares the output of the register 66, that is, the lower 2 bits C of (A + B) with the output of the register 62, that is, the dither conformance G of 3 bits, and G
If is larger than C, a flag "0" indicating that dither processing should be performed is output, and if G is C or less, a flag "1" indicating that binary processing is performed is output.

Instead of adopting the configuration of FIG. 2, a matrix in which 0 to 3 are marked so that the same numbers are not fixed may be used.

In the embodiment of the image processing apparatus shown in FIG.
When the goodness of fit for halftone processing, that is, dithering, is the maximum value, dither processing is selected, when the goodness of fit is the minimum value, binary processing is selected, and when the goodness of fit has a value between the maximum value and the minimum value, Since the dither processing and the binary processing are selected at a frequency according to the goodness of fit, it is possible to perform processing having an intermediate property between the dither processing and the binary processing according to the goodness of fit. It is possible to improve the image quality of an image in which it is impossible to uniformly determine which process is suitable only with pixel data. Further, even if the fitness changes, it does not affect all the pixels that make up the image. Therefore, compared to the method in which the fitness change affects the thresholds of all pixels, the density changes gently. The image quality does not change suddenly.

In the embodiment of FIG. 1, the fitness for dither processing is obtained, but a block-unit binary processing fitness calculating device for finding fitness for binary processing may be provided. In this case, the binary dither selection flag calculating device (processing method selecting means) selects binary processing when the goodness of fit is the maximum value, and selects halftone processing when the goodness of fit is the minimum value. When the degree takes a value between the maximum and minimum values,
The binary processing and the halftone processing may be selected at a frequency according to the degree of compatibility. With this configuration, the binary processing is selected when the goodness of fit to the binary processing is the maximum value, and the halftone processing is selected when the goodness of fit is the minimum value.
When the goodness of fit takes a value between the maximum value and the minimum value, the binary processing and the halftone processing are selected with a frequency according to the goodness of fit. Since it is possible to perform processing having an intermediate characteristic of, it is possible to improve the image quality of an image in which it is impossible to uniformly determine which processing is suitable with only pixel data in a block. Further, even if the fitness changes, it does not affect all the pixels that make up the image. Therefore, compared to the method in which the fitness change affects the thresholds of all pixels, the density changes gently. The image quality does not change suddenly.

FIG. 6 shows the configuration of the second embodiment of the image processing apparatus of the present invention. The image data capturing device 1 supplies the image count to the dither threshold calculation device 2 and
The image data is supplied to the block-based dither compatibility calculation device (compatibility calculation means) 3 and the output determination device 10. The dither threshold calculation device 8 calculates the dither processing threshold D from the image count and supplies it to the threshold calculation device 9. The block unit dither suitability calculation device 3 divides all pixels forming an image into blocks of 8 pixels × 1 column as shown in FIG. 3, and includes data of 8 pixels in the processing target block or surrounding blocks. The feature amount of the processing target block such as the maximum edge and the number of pulses is calculated from the data of 32 pixels, fuzzy inference is performed using the feature amount as an input, the fitness G of the processing target block to the dither processing is calculated, and the fitness G is calculated. It is expressed in 5 values and output. The configuration and operation of the block-based dither compatibility calculation device 3 are the same as in the embodiment of FIG.

The threshold value calculation device 9 receives the binary processing threshold value T, the dither processing threshold value D and the dither matching degree G, and performs the processing shown in FIG. 7 to calculate the actual threshold value TR. Threshold calculation device 9
First, in step S1, the following equation 1 is calculated to calculate the threshold T1. T1 = T × (1-G / 4) (Equation 1)

Next, the threshold value calculation device 9 calculates the threshold value T2 by performing the calculation of the following expression 2 in step S2. T2 = T + (255-T) × G / 4 (Equation 2)

Therefore, the threshold values T1 and T2 take the following values (provided that T = 95) according to the goodness of fit G:

Next, the threshold value calculation device 9 compares D with T1 in step S3, and if T1 is greater than or equal to D, then T
1 is set as the actual threshold value TR (step S4). T1 is D
When it is smaller, the threshold calculation device 9 determines in step S5.
In, comparing D and T2, and if D is T2 or more,
Let T2 be the actual threshold value TR (step S6). If D is smaller than T2, the threshold calculation device 9 sets D as the actual threshold TR (step S7). The output determination device 10 is
The actual threshold value TR output from the threshold value calculation device 9 is received to determine the black and white of the pixel.

Therefore, in the embodiment shown in FIG. 6, when the goodness-of-fit G is 0, the binary processing is completed and the goodness-of-fit G is 4
In the case of (normalized to 1), the dither processing is completely performed. When the goodness-of-fit G has an intermediate value, the intermediate-luminance pixels are dithered and the high-luminance pixels and the low-luminance pixels are binarized, so that an image with clear edges is obtained.

In the embodiment of FIG. 6, two threshold values T1 and T2 are used, but either one may be used. In this case, white background emphasis and black background emphasis may be used properly.

Further, instead of the threshold value calculating device 9 of FIG.
A constant determining means may be provided for determining the actual threshold value TR of the dither matrix according to the following expression 3 according to the dither compatibility G. However, this is limited to the case where γ correction is not applied to the dither matrix and the threshold value is simple (Equation 3
, T is the threshold value for binary processing, and D is the threshold value for dithering the target pixel). TR = (D−T) × G / 4 + T (Formula 3)

By doing so, the dither (halftone) processing threshold value and the binary processing threshold value are weighted and averaged according to the dither matching degree G of the block, and the weighted average value is used as the threshold value for determining black and white. Therefore, it is possible to perform processing having an intermediate property between dither processing and binary processing according to the degree of conformity, and it is not possible to uniformly determine which processing is suitable only with pixel data in a block. The image quality of the image can be improved.

FIG. 8 shows the configuration of the third embodiment of the image processing apparatus of the present invention. In this example, the dither matrix has γ
This is a preferred embodiment when correction is applied and the threshold value is complicated. The image data capturing device 1 supplies the image count to the dither threshold dither number calculating device 11 and also supplies the image data to the block unit dither compatibility calculating device (compatibility calculating means) 3 and the output determining device 10. The dither threshold dither number calculation device 11 calculates the dither processing threshold D and the dither number N from the image count, and supplies the dither processing threshold D and the dither number N to the threshold calculation device 12. The block unit dither compatibility calculation device 3 is the same as that of the embodiment of FIG.

The threshold value calculation device 12 receives the binary processing threshold value T, the dither processing threshold value D, the dither number N of the target pixel, the binary dither number and the dither matching degree G, and performs the processing of FIG. The actual threshold value TR is calculated. The threshold value calculation device 11 first calculates the conversion dither number NR by calculating the following Expression 4 in step S11 (here, the binary dither number is 39). NR = (N-39) × G / 4 + 39 (Formula 4)

Next, the threshold value calculation device 11 performs step S1.
In step 2, the following equation 5 is calculated to calculate the actual dither threshold value TR of the conversion dither number NR. TR = D [NR] (Equation 5)

The output determining device 10 receives the actual threshold value TR output from the threshold value calculating device 9 and determines the black and white of the pixel.

According to the embodiment shown in FIG. 8, it is possible to perform processing having an intermediate characteristic between dither processing and binary processing according to the goodness of fit G. It is possible to improve the image quality of an image for which it is not possible to uniformly determine whether is suitable.

Instead of changing the threshold value of the dither matrix, the size of the dither matrix may be changed as follows according to the dither processing suitability G. Goodness of fit 0 1 × 1 Goodness of fit 1 2 × 2 Goodness of fit 2 4 × 4 Goodness of fit 3 6 × 6 Goodness of fit 4 8 × 8

The smaller the size of the dither matrix, the smaller the resolution. Therefore, since it is possible to perform processing having an intermediate characteristic between other image processing methods such as dither processing and binary processing according to the goodness of fit G, which image processing method is suitable only with pixel data in a block. It is possible to improve the image quality of an image that cannot be uniformly determined. Further, even if the goodness of fit G takes an intermediate value, since a correct dither matrix is used as the distribution of the threshold value, unnatural states such as blown-out highlights and crushed blacks are reduced.

FIG. 10 shows the configuration of the fourth embodiment of the image processing apparatus of the present invention. The image data capturing device 1 supplies the image count to the dither number calculating device 2 and
The image data is supplied to the block unit character conformity calculating device (conformance calculating unit) 14, the fixed binary processing device 15, the floating binary processing device 16 and the image data memory 17. The dither number calculation device 2 calculates the dither number from the image count and supplies it to the fixed / floating selection flag calculation device (processing method selection means) 18. The fixed binary processing device 15 supplies the fixed binary processing result for the image data to the output determination device 19. The floating binary processing device 16 outputs the floating binary processing result for the image data, and determines the output device 19.
Supply to.

As shown in FIG. 11, when the surrounding pixel data stored in the image data memory 17 are A, B and C and the pixel data to be processed is D, the floating binary processor 16 will be described next. Equations 6 and 7 are calculated to obtain the floating binary threshold value T. S = A + B + C (Equation 6) T = S / 4 + 24 (32) (Equation 7) If D is larger than T, the pixel to be processed is determined to be white, and if D is T or less, the pixel to be processed is black. decide. Such a processing result is supplied to the output determination device 19.

The block unit character conformance calculation device 14
All pixels forming the image are divided into blocks of 8 pixels x 1 column as shown in Fig. 3, and the maximum edge and the number of pulses are calculated from the data of 8 pixels in the block to be processed or the data of 32 pixels including the surrounding blocks. The feature amount of the block to be processed is calculated, and fuzzy inference is performed using the feature amount as an input to obtain the degree of conformance G to the character area of the block to be processed. Here, the goodness of fit G is represented by five values as follows. Goodness of fit = {0.00, 0.25, 0.50, 0.75, 1.00} However, in actual hardware, goodness of fit = {0, 1,
2, 3, 4} may be used.

The fixed / floating selection flag calculator 18 receives the dither number from the dither number calculator 2 and
From the block unit character conformance calculation device 14, the character conformance G
In response to this, a selection flag is output to the output determination device 19 to cause the fixed binary processing result and the floating binary processing result to be selected at a frequency according to the following goodness of fit G. Goodness of fit 0 All Fixed Binary processing Goodness of fit 1 4 times Floating binary processing Goodness of fit 2 4 times 2 Floating binary processing Goodness of fit 3 4 3 times Floating binary processing Goodness of fit 4 All floating 2 Value processing

As described above, since the dither matrix has a period of 4, if the remainder of 4 is processed as it is, the area to be binary-processed and the area to be dithered will be separated. Therefore, it is preferable to change the cycle. FIG. 12 shows an example of the fixed / floating selection flag calculation device 18 configured in consideration of this point. The register 181 holds a 6-bit dither number N. Register 182 is 3
The bit character conformance G is held, and the 3-bit comparator 187
Output to. The register 183 receives the output of the register 181, and holds the middle 2 bits B of the dither number N.
The register 184 receives the output of the register 181 and holds the lower 2 bits A of the dither number N. The 2-bit adder 185 adds the contents of the registers 183 and 184. However, it is not carried forward. The register 186 holds the output of the adder 185, that is, the lower 2 bits C of (A + B). The register 186 always holds 0 as the most significant bit. The comparator 187 outputs the output of the register 186, that is, the lower 2 bits C of (A + B).
Is compared with the output of the register 182, that is, the 3-bit character conformance G, and if G is larger than C, a flag "0" indicating that fixed binary processing should be performed is output, and if G is C or less, , A flag “1” indicating that floating binary processing should be performed is output.

Fixed / floating selection flag calculation device 1 of FIG.
8 determined the selection flag based on the dither number,
The selection flag may be determined based on the image count instead of the dither number.

When the floating binary processing is performed on the character image, the line is clearly visible, but on the other hand, when the background is noisy like a newspaper, there is a drawback that dots are left on the background. is there. On the other hand, in the fixed binary processing, the line is thin, but the ground is clean. In the embodiment of FIG. 10, the floating binary process is selected when the goodness of fit G for the character area is the maximum value, and the fixed binary process is selected when the goodness of fit G is the minimum value. Floating processing can be performed, and fixed binary processing can be performed on the background portion, so that a clear character image can be obtained. When the goodness of fit G takes a value between the maximum value and the minimum value, the floating binary process and the fixed binary process are selected at a frequency according to the goodness of fit G. Therefore, since it is possible to perform processing having an intermediate property between floating binary processing fixed and binary processing according to the goodness of fit G, it is uniformly determined which processing is suitable only with pixel data in a block. It is possible to improve the quality of images that cannot be displayed. Further, even if the fitness changes, it does not affect all the pixels that make up the image. Therefore, compared to the method in which the fitness change affects the thresholds of all pixels, the density changes gently. Image quality does not change suddenly

FIG. 13 shows the configuration of the fifth embodiment of the image processing apparatus of the present invention. The image data capturing device 1 supplies the image data to the image data memory 20, the floating binary threshold value calculating device 21, the block unit character conformance calculating device (conformance calculating means) 14 and the output determining device 23. The floating binary threshold value calculation device 21 reads out the surrounding pixel data of the processing target pixel from the image data memory 20,
The floating binary processing threshold value T of the pixel to be processed is calculated based on the above, and is supplied to the threshold value calculation device 22. The block-unit character conformance calculation device 14 is the same as that of the embodiment shown in FIG.

The threshold value calculation device 22 receives the floating binary processing threshold value T, the fixed binary processing threshold value D, and the character matching degree G, and
The following equation 8 is calculated to calculate the actual threshold value TR. TR = (D−T) × G / 4 + T (Equation 8)

In the embodiment shown in FIG. 13, a weighted average of the fixed binary processing threshold D and the floating binary processing threshold T is obtained according to the goodness of fit G of the block to the character area.
Since this weighted average value is used as the threshold value TR for the pixel to be processed, it is possible to perform processing having an intermediate characteristic of fixed binary processing and floating binary processing according to the goodness of fit G, so that the pixels in the block It is possible to improve the image quality of an image in which it is impossible to uniformly determine which process is suitable with only data.

When a pixel is painted black by image processing, the logical product of the binary processing result and the dither processing result is taken to emphasize white, and the logical sum is taken to emphasize black. Depending on the type and use of the image, white enhancement may be desired or black enhancement may be desired. For example, if you want to make it easier to see the black characters in the photo, but you want to dither the photo itself, black emphasis is appropriate, and you want to make the white edges of the photo clear.
White emphasis is suitable.

FIG. 14 shows the configuration of the sixth embodiment of the image processing apparatus of the present invention. This embodiment is suitable for emphasizing only the white edge portion of a (halftone dot) photographic image.
The image data capturing device 1 supplies the image data to the block unit density difference calculating device (density difference calculating means) 24, the binary processing device 4 and the dither processing device 5. The binary processing device 4 supplies the binary processing result for the image data to the output determining device 26. The dither processing device 5 supplies the dither processing result for the image data to the output determining device 26.

The block unit density difference calculation device 24 divides all the pixels forming the image into blocks of 8 pixels × 1 column as shown in FIG. 3, and calculates the density difference in the data of 8 pixels in the processing target block. It is calculated and output to the comparator 25. Comparator 2
5 outputs the selection flag “1” to the output determining device 26 when the density difference is larger than the threshold Th. Accordingly
The output determining device 26 outputs a logical product of the binary processing result and the dither processing result, that is, min (binary processing result, dither processing result). Further, the comparator 25 determines that the density difference is the threshold value T.
When it is less than or equal to h, the selection flag “0” is output to the output determining device 26. In response to this, the output determination device 26 outputs the dither processing result. Therefore, for example, only the white edge portion of the photographic image can be emphasized.

FIG. 15 shows the configuration of the seventh embodiment of the image processing apparatus of the present invention. This embodiment is also an embodiment suitable for emphasizing only the white edge portion of the (halftone dot) photographic image.
The image data capturing device 1 supplies the image count to the dither number calculating device 2, and at the same time, stores the image data in the block unit white enhancement conformance calculating device (conformance calculating means) 27,
It is supplied to the binary processing device 4 and the dither processing device 5. The dither number calculation device 2 calculates the dither number from the image count and supplies it to the binary dither selection flag calculation device (processing method selection means) 28. The binary processing device 4 supplies the binary processing result for the image data to the AND circuit 29.
The dither processing device 5 supplies the dither processing result for the image data to the AND circuit 29 and outputs the output determining device 3.
Supply to 0. The logical product circuit 29 supplies the logical product of the binary processing result and the dither processing result to the output determining device 30.

Block unit white enhancement suitability calculating device 27
Is 8 pixels x 8 pixels as shown in FIG.
It is divided into blocks of one column, and the feature amount of the process target block such as the maximum edge and the number of pulses is calculated from the data of 8 pixels in the process target block or the data of 32 pixels including the surrounding blocks, and the feature amount is input. Fuzzy inference is performed to find the suitability of the target block for white enhancement processing. Here, the goodness of fit is represented by five values as follows. Goodness of fit = {0.00, 0.25, 0.50, 0.75, 1.00} However, in actual hardware, goodness of fit = {0, 1,
2, 3, 4} may be used.

The binary dither selection flag calculation device 28 receives the dither number from the dither number calculation device 2 and
Upon receiving the dither suitability from the block unit white enhancement suitability calculating device 27, the selection flag is output to the output determining device 30,
The dither processing result and the logical product of the binary processing result and the dither processing result are selected at a frequency according to the following conformity. Goodness of fit 0 All dithering Goodness of fit 1 4 times 1 Logical product goodness of fit 2 4 4 times Logical product goodness of fit 3 4 3 times Logical product goodness of fit 4 All logical products

In the embodiment of FIG. 15, the logical product is selected when the degree of conformity to the white enhancement processing is the maximum value, and the dither processing result is selected when the degree of conformance is the minimum value. When taking a value between the values, the dither processing result and the logical product of the binary processing result and the dither processing result are selected at a frequency according to the goodness of fit, and accordingly, the dithering process is performed according to the goodness of fit. Since it is possible to perform a process having an intermediate characteristic of the white enhancement process, it is possible to improve the image quality of an image in which it is impossible to uniformly determine which process is suitable only with the pixel data in the block. Further, even if the fitness changes, it does not affect all the pixels that make up the image. Therefore, compared to the method in which the fitness change affects the thresholds of all pixels, the density changes gently. The image quality does not change suddenly.

In the embodiment of FIG. 15, the dither processing result and the logical product of the binary processing result and the dither processing result are selected at a frequency according to the degree of conformity, so that the dither processing result and the logical product are obtained. Mixed, but instead of doing this
A constant determining unit may be provided to determine the actual threshold value TR of the dither matrix according to the following Expression 9 according to the white enhancement conformance G. However, this is limited to the case where the dither matrix is not subjected to γ correction and the threshold value is simple. (In Expression 9, T is a threshold for white enhancement processing, and D is a threshold for dither processing of the target pixel.) TR = (D−T) × G / 4 + T (Expression 9)

[0091]

According to the first image processing apparatus of the present invention,
Divide the image into blocks containing a certain number of pixels
The block's goodness of fit is calculated, and the goodness of fit is the maximum.
When halftone processing is selected, and when the goodness of fit is the minimum value
Is binary processing, and the goodness of fit is between the maximum and minimum values
When taking, either halftone processing or binary processing
Since I chose to select the frequency according to the fitness,
Which process is suitable only for pixel data in
It is possible to improve the quality of images that cannot be judged uniformly.
Wear. In addition, even if the degree of conformity changes, all images that make up the image
Since there is no influence on the element, the change in fitness is applied to all pixels.
Compared with the method that affects the threshold of
The image quality does not change suddenly in the changing part.

According to the second image processing apparatus of the present invention,
Divide the image into blocks containing a certain number of pixels for binary processing
Calculates the goodness of fit of the block for the
Select binary processing, and when the goodness of fit is the minimum value, it is intermediate
Select the adjustment process, and set the goodness of fit to a value between the maximum and minimum values.
When processing, adapt either binary processing or halftone processing
Since I chose to select the frequency according to the degree,
Which processing is suitable for the pixel data of
You can improve the quality of images that cannot be judged evenly.
It In addition, even if the fitness changes, all pixels that make up the image
Since it does not affect the
Compared with the method that affects the threshold, the concentration changes slowly.
The image quality does not change suddenly in the changing part.

The third image processing apparatus and the first embodiment of the present invention
According to the image processing method, the image is made to include a certain number of pixels.
And the block's suitability for halftone processing.
Calculate, calculate the threshold value according to the calculated conformance, and calculate
And the threshold value for halftone processing
Used to determine the threshold for black and white decisions
Therefore, halftone processing is performed for pixels with intermediate brightness, and high
Perform binary processing on luminance and low luminance pixels
You can get a sharp-edged image
It

According to the fourth image processing apparatus of the present invention,
An image consisting of several pixels into a block containing a certain number of pixels
Divide, calculate the degree of conformity of the block to the character area,
Select the floating binary processing when the goodness of fit is the maximum
Is the minimum value, fixed binary processing is selected and the goodness of fit is maximized.
When taking a value between a large value and a minimum value,
Select either fixed binary processing at a frequency according to the goodness of fit
Since it was done, only the pixel data in the block
Is an image that cannot be uniformly determined which processing is suitable.
The image quality of the image can be improved. Also, the fitness may change
Even if it is converted, it may affect all the pixels that make up the image.
Since there is no such change, the change in fitness affects the thresholds of all pixels.
Compared with the method, the image quality is sharper in the part where the density changes gently.
It doesn't change.

The fifth image processing apparatus and the second image processing apparatus of the present invention
According to the image processing method, a predetermined number of images including a plurality of pixels
It is divided into blocks containing
Calculation of the goodness of fit, fixed binary threshold and floating binary
The processing threshold and the weighted average are calculated according to the goodness of fit.
Use the found average value as the threshold for the target pixel
Fixed binary processing and floating binary processing according to the goodness of fit.
Because it is possible to perform processing with intermediate properties of
Which processing is suitable only for pixel data in a block
To improve the quality of images that cannot be uniformly determined.
You can

[0096]

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103]

[0104]

[0105]

[0106]

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a first embodiment of an image processing apparatus of the present invention.

FIG. 2 is a block diagram showing a configuration example of a binary dither selection flag calculation device in FIG.

FIG. 3 is an explanatory diagram showing an example of a pixel block.

FIG. 4 is an explanatory diagram showing an example of a dither matrix.

FIG. 5 is an explanatory diagram showing a relationship between a dither number, lower 2 bits A of the dither number, middle 2 bits B of the dither number, and lower 2 bits of (A + B).

FIG. 6 is a block diagram showing the configuration of a second embodiment of the image processing apparatus of the present invention.

7 is a flowchart showing the operation of the threshold value calculation device 9 of FIG.

FIG. 8 is a block diagram showing the configuration of a third embodiment of the image processing apparatus of the present invention.

9 is a flowchart showing an operation of the threshold value calculation device 12 of FIG.

FIG. 10 is a block diagram showing the configuration of a fourth embodiment of the image processing apparatus of the present invention.

FIG. 11 is an explanatory diagram showing a relationship between a floating binary processing target pixel and surrounding pixels.

12 is a block diagram showing a configuration example of a fixed / floating selection flag calculation device in FIG.

FIG. 13 is a block diagram showing the configuration of a fifth embodiment of the image processing apparatus of the present invention.

FIG. 14 is a block diagram showing the configuration of a sixth embodiment of the image processing apparatus of the present invention.

FIG. 15 is a block diagram showing the configuration of a seventh embodiment of the image processing apparatus of the present invention.

[Explanation of symbols]

3 block unit dither compatibility calculation device 4 binary processing device 5 dither processing device 6, 28 binary dither selection flag calculation device 7, 10, 13, 19, 23, 26, 30 output determination device 8 dither threshold calculation device 9, 12, 22 Threshold Value Calculation Device 15 Fixed Binary Processing Device 16 Floating Binary Processing Device 18 Fixed / Floating Selection Flag Calculation Device 24 Block Unit Density Difference Calculation Device 25 Comparator 27 Block Unit White Strong Adaptation Calculation Device

Continuation of the front page (72) Hajime Okumura Hajime Todocho, Hanazono-cho, Ukyo-ku, Kyoto City, Kyoto Prefecture Omron Co., Ltd. (72) Yoshinori Yamaguchi 10th Hanazono-Tudocho, Ukyo-ku, Kyoto City, Kyoto Prefecture (72) ) Osamu Oshima, Omron Co., Ltd., 10 Hodoen-cho, Uenyo-ku, Kyoto City, Kyoto Prefecture (56) References JP-A-58-10961 (JP, A) JP-A-2-84879 (JP, A) JP-A-SHO 63-59272 (JP, A) JP-A 61-288561 (JP, A) JP-A 63-71782 (JP, A) JP-A 63-71781 (JP, A) JP-A 61-98069 (JP, A) A) JP 1-302961 (JP, A) JP 1-181380 (JP, A) JP 2-100575 (JP, A) JP 62-18171 (JP, A) (58) Survey Areas (Int.Cl. ⁷ , DB name) H04N 1/40-1/409 H04N 1/46 H04N 1/60

Claims (7)

(57) [Claims] 1. An image processing apparatus for determining black and white of each pixel of an image composed of a plurality of pixels, comprising means for dividing the image into blocks including a predetermined number of pixels, and a halftone corresponding to the type of the image. Of a plurality of image processing methods including image processing and binary processing, and a matching degree calculation unit that calculates a matching degree of the block with respect to the halftone processing, and the matching from the plurality of image processing methods. and a processing method selecting means for selecting a processing method corresponding to time, the processing method selection unit, when the matching degree is equal to the maximum value
When the halftone processing is selected and the goodness of fit is the minimum value,
Selects the binary processing, and the goodness of fit is equal to or smaller than the maximum value.
When taking a value between the minimum values,
An image processing apparatus , wherein the binary processing is selected at a frequency according to the compatibility . 2. A monochrome image of each pixel of an image composed of a plurality of pixels
An image processing device for determining Means for dividing the image into blocks containing a predetermined number of pixels
When, Multi-level processing including halftone processing and binary processing depending on the image type
Number of image processing methods for the binary processing
A compatibility calculation means for calculating the compatibility of the lock, According to the degree of conformity among the plurality of image processing methods
A processing method selecting means for selecting a processing method, The processing method selecting means is configured to detect when the conformity is the maximum value.
When the binary processing is selected and the goodness of fit is the minimum value,
Select the halftone process and set the goodness of fit to the maximum value and the previous value.
When taking a value between the minimum values, the binary processing and the
To select halftone processing at a frequency according to the degree of conformity
An image processing device characterized by: 3. An image processing apparatus for determining black and white of each pixel of an image consisting of a plurality of pixels, said means for dividing the image into blocks including a predetermined number of pixels, and binary processing and halftone processing. among the fitness calculating means for calculating a goodness of fit of the blocks with respect to the halftone processing, according to the matching degree calculated by the matching degree calculation means
A threshold value calculating means for calculating a threshold value, and the degree of conformity calculated by the threshold value calculating means.
Using the threshold and the threshold of the halftone processing, the white
An image processing apparatus, comprising: a threshold value determining unit that determines a threshold value for determining black . 4. A means for dividing an image composed of a plurality of pixels into blocks containing a predetermined number of pixels, and a fitness calculation means for computing the fitness of the block to the character area among the character area and the background area. A processing method selecting means for selecting either fixed binary processing or floating binary processing according to the goodness of fit , wherein the processing method selecting means is used when the goodness of fit is a maximum value.
When the floating binary processing is selected and the goodness of fit is the minimum value
Select the fixed binary processing for the
The floating binary value when taking a value between the value and the minimum value.
Processing and the fixed binary processing at a frequency according to the goodness of fit
An image processing device characterized by selecting . 5. An image processing apparatus comprising fixed binary processing means for performing fixed binary processing on pixels, and floating binary processing means for performing floating binary processing on pixels, comprising a plurality of pixels. Means for dividing an image containing a predetermined number of pixels into a block, a fitness calculation means for computing the fitness of the block with respect to a character region, the fixed binary processing threshold and the floating binary processing threshold And a threshold value determining unit that uses the weighted average value as a threshold value for the pixel to be processed. 6. An image processing method for determining black and white of each pixel of an image composed of a plurality of pixels, the method comprising a step of dividing the image into blocks including a predetermined number of pixels, and a binary processing or a halftone processing. Among these, the fitness calculation step of calculating the fitness of the block with respect to the halftone processing, and the fitness calculated by the processing of the fitness calculation step.
A threshold value calculation step of calculating a threshold value according to the degree of conformity, and the adaptation calculated by the processing of the threshold value calculation step
Using the threshold according to the degree and the threshold of the halftone processing
The threshold determination step for determining the threshold for determining the black and white.
An image processing method characterized by comprising a-up. 7. An image processing method in an image processing apparatus having fixed binary processing means for performing fixed binary processing on pixels, and floating binary processing means for performing floating binary processing on pixels. A dividing step of dividing an image including a plurality of pixels into blocks including a predetermined number of pixels, a matching degree calculating step of calculating a matching degree of the block with respect to a character region, a threshold of the fixed binary processing, and the floating binary An image processing method, comprising: a weighting average of a threshold value processing and a threshold value determination step of setting the weighted average value as a threshold value for a pixel to be processed.