JPH11220626A - Method for forming image data and device therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress the deterioration of image quality of a recorded image by satisfactorily holding a dot shape. SOLUTION: Input image data SP for a picture element P under consideration are turned into error correction data SP' when the accumulated values of errors generated from picture elements processed in advance are added. The error correction data SP' are binarized by a binarizing part 14 based on a threshold Tp obtained from a threshold memory 13. An error detecting part 15 transmits an error ep generated by binarization to an error distribution part 17. On the other hand, based on the position information of the picture element P under consideration and binary data Bp , a distribution ratio memory 16 transmits distribution ratios W0 -W11 of the error ep for preliminarily stored non-processed picture elements P0 -P11 positioned adjacently to the picture element P under consideration to the error distributing part 17. The error distributing part 17 distributes the error ep to each non-processed picture element P0 -P11 according to the distribution ratios W0 -W11 obtained from the distribution ratio memory 16.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image data forming method for forming one halftone dot by a set of a plurality of pixels and changing the size of the halftone dot to reproduce pseudo multiple gradations. Regarding the device.

[0002]

2. Description of the Related Art Conventionally, when a color image including a plurality of color components is printed, halftone dots are recorded for each color component. In halftone dot recording, it is necessary to reproduce multiple tones in a pseudo manner in order to maintain consistency with the original image. As a conventional halftone dot forming method for reproducing such a pseudo multi-tone, the size of one halftone dot is changed in accordance with the original image, and the halftone dot is simulated by a set of large and small halftone dots. There is a method for reproducing multiple gradations. In prepress printing,
An original image expressed in multiple gradations is reproduced in a pseudo manner by a set of halftone dots having different sizes such as large and small.

When a recorded image is generated by forming such halftone dots, the halftone patterns for each color component are arranged at different angles in order to reduce moiré caused by overlapping the halftone patterns for each color component. Is done. For example, Y (yellow), M (magenta), C
For each of the color components (cyan) and K (black), angles of approximately 0 °, 15 °, 75 °, and 45 ° are selected.

As an example, a method for forming a halftone image of about 15 ° will be described. FIG. 10 is a diagram for explaining a method of forming a halftone dot pattern of about 18.435 ° (= tan ⁻¹ ３). In FIG. 10, a scale is shown for each recording pixel, and coordinate axes Fx and Fy are set on a recording material (recording paper, recording film, or the like). The halftone dot region DR is inclined by about 18.435 ° with respect to the coordinate axes Fx and Fy.

As shown in FIG. 10, halftone dots of the same pattern are repeatedly generated in each of N × N (where N is an arbitrary integer) repeating areas AR. The repeating area AR includes a plurality of halftone dot areas DR in which one halftone dot is formed.

[0006] Then, a threshold value is set for each recording pixel of the repetition area AR, and is binarized to 1 or 0 by comparing with the input image data. In each halftone dot region DR included in the repetition region AR, a small threshold value is stored in the center portion, and a larger threshold value is stored in a spiral shape as the distance from the center portion increases. Accordingly, the pixels are sequentially blackened for each pixel from the center of the halftone dot region DR, and the plurality of blackened pixels form one unit. This one unit is 1
Individual dots.

The threshold value for each recording pixel existing in the repetition area AR is calculated by a computer and stored in advance in a memory. The threshold value corresponding to each recording pixel in the repeating area AR stored in the memory is set so that the position of the recording pixel on the coordinate axes Fx and Fy can be read as an address.

Accordingly, an arbitrary position (fx, f
The threshold value of the recording pixel in y) is obtained by dividing the position fx on the Fx axis by the size N of the repetition area AR as an address on the Fx axis, and using the position fy on the Fy axis as the repetition area A
The remainder obtained by dividing by the size N of R can be obtained as an address on the Fy axis by accessing the memory. In this way, the threshold value set for the repetition area AR is repeatedly applied to all the areas of the recording material. Then, by performing binarization processing based on the threshold value for all the pixels, a recorded image is formed by halftone dots having different sizes according to the original image.

Here, it is assumed that the dynamic range of each pixel of the original image is 256 gradations of 0 to 255, and as an example, the input image data for all the pixels of the original image is "128". In this case, when an image is recorded by the dot formation, a recorded image as shown in FIG. 11 is obtained. In FIG. 11, the input image data is "128".
Therefore, the portion where the dot recording is performed (black portion) and the portion where the recording is not performed are distributed almost equally.

As described above, when the original image is printed, if halftone dots are formed, by changing the size of one halftone dot in accordance with the original image, the multi-gradation of the original image can be simulated. Since it can be reproduced, consistency with the original image is maintained.

[0011]

However, when the above-described halftone dot formation is performed, the shape of the halftone dot is not maintained well,
There is a problem that moire may occur and image quality may be degraded.

For example, FIG. 12 shows a dot image of a striped image in which fine lines are arranged in the horizontal direction. In FIG. 12, the input image data for the thin line portion is “128”, and the input image data for the background portion is “0”.

In a recorded image as shown in FIG.
The original image and a threshold value for forming a halftone dot interfere with each other, and moire fringes are generated from the upper right portion to the lower left portion separately from the original horizontal stripes of the original image, resulting in deterioration of image quality.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and provides an image data forming method and apparatus capable of suppressing deterioration in image quality of a recorded image by maintaining a good dot shape. The purpose is to do.

[0015]

In order to achieve the above object, according to the present invention, the multi-valued input image data inputted for each pixel is subjected to a binarization process to be binarized. An image data forming method for generating image data, comprising: (a) 2
Generating error addition data by adding a cumulative value of an error generated from a pixel that has already been binarized to input image data of a target pixel to be valued;
(b) performing a binarization process on the target pixel by comparing a threshold value for forming a halftone dot corresponding to the position of the target pixel with error addition data; and (c) a binarization process on the target pixel. Detecting the error caused by the
(d) When the result of the binarization process for the target pixel is “1” as a logical value, the threshold value corresponding to each pixel is larger for a plurality of pixels included in the predetermined area in descending order. , A step of distributing more errors, and (e) when a result of the binarization processing for the target pixel is “0” as a logical value, a plurality of pixels included in a predetermined area are And distributing more errors in ascending order of the threshold value corresponding to each pixel.

According to a second aspect of the present invention, in the image data forming method according to the first aspect, the step (d) comprises the step of: (d-1) determining that a result of the binarization processing on the target pixel is a logical value. In the case of "1", for a plurality of pixels included in the predetermined region, the image data of each pixel includes a step of distributing more errors in ascending order, and the step (e) includes: , (E-1), when the result of the binarization process for the target pixel is “0” as a logical value, the image data of each pixel is determined for a plurality of pixels included in the predetermined area. Is characterized by having a step of further distributing more errors in descending order of.

According to a third aspect of the present invention, there is provided an image data forming apparatus for performing binarization processing on multi-valued input image data input for each pixel to generate binarized image data. a) adding means for generating error addition data by adding the cumulative value of errors generated from the pixels already subjected to the binarization processing to the input image data of the target pixel to be binarized; b) binarization means for performing binarization processing on the pixel of interest by comparing a threshold value for forming a halftone dot corresponding to the position of the pixel of interest with error addition data; Error detection means for detecting an error generated by the binarization processing; and (d) when a result of the binarization processing for the target pixel is “1” as a logical value, Threshold value corresponding to each pixel Are set in the descending order, and when the result of the binarization process for the pixel of interest is “0” as a logical value, a plurality of pixels included in a predetermined region are A distribution ratio setting unit that sets the distribution ratio of the error to be large in the order of smaller threshold values corresponding to the respective pixels; and (e) a plurality of distribution regions included in a predetermined area based on the distribution ratio obtained by the distribution ratio setting unit. Error distributing means for distributing an error to the pixels.

According to a fourth aspect of the present invention, in the image data forming apparatus according to the third aspect, the distribution ratio setting means includes (d)
-1) When the result of the binarization process for the target pixel is “1” as a logical value, the image data of each pixel is determined in ascending order for a plurality of pixels included in the predetermined area. In addition, when the distribution ratio of the error is set to a larger value and the result of the binarization processing for the pixel of interest is “0” as a logical value, a plurality of pixels included in a predetermined region are Means for setting the error distribution rate to be larger in the order of larger image data of the pixels.

In the present invention, the logical values "1" and "0" are relative, and the actual signal level depends on whether a positive expression or a negative expression is used in image recording or image display. Is determined corresponding to "1" and "0". Also, the magnitude relationship between the threshold values is determined by the absolute value.

In the following embodiment, a positive expression is taken as an example, and "1" is used as a signal value and "1" is used as a logical value.
Matches.

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENTS <1. Overview of Apparatus> FIG. 1 is a schematic diagram of a configuration of an image processing system to which an embodiment of the present invention is applied. The image signal supply device 100 is a device that can output an image signal from an image reading device such as a scanner or an image signal storage device such as an optical disk or a magnetic disk. The pre-processing unit 200 is a device that performs various processes on an image signal provided from the image signal supply device 100. The various processes include, for example, a tone conversion process and a sharpness enhancement process.

The signals subjected to various types of image processing in the pre-processing unit 200 are input to the image data forming apparatus 300 as multi-value (multi-tone) input image data for each pixel. Further, in addition to the input image data of the pixel of interest, which is the pixel to be processed, positional information on the coordinate axis of the pixel of interest is input to the image data forming apparatus 300.

The image data forming apparatus 300 is a part to which the present invention is applied. The image data forming apparatus 300 performs a binarization process on multi-valued input image data for each pixel provided from the preprocessing unit 200 to perform a binary process. Generate recording image data. The recording image data for each pixel generated by the image data forming device 300 is sent to the image recording device 400.

The image recording apparatus 400 records an image on a predetermined recording material based on recording image data sent from the image data forming apparatus 300.

Before describing the image data forming apparatus 300 to which the present invention is applied, an image processing mode will be described. FIG. 2 is a diagram showing an image composed of pixels in p rows and q columns. When processing an image composed of pixels in p rows and q columns as shown in FIG. 2, a pixel to be processed (hereinafter, referred to as a pixel to be processed)
P is referred to as a “pixel of interest”) from the pixel at row 0 and column 0, and then the pixel at row 0 and column 1 is processed. When the processing is completed up to the 0th row and the qth column, the process returns to the 1st row and the 0th column, and the same processing is repeated. When the processing is completed up to the pth row and the qth column, the processing for this image is completed. Will be. As described above, the second image of the image performed by the image data forming apparatus 300 is
The binarization processing is a processing mode in which the pixel of interest P is performed one pixel at a time. Then, similar processing is performed on a plurality of color components (Y, M, C, K) obtained from the original image. FIG. 2 shows a state in which the target pixel P has advanced to the i-th row and the j-th column, and the hatched area shown in the figure is the area where the binarization processing has been completed.

In the following embodiment, it is assumed that the input image data input to the image data forming apparatus 300 has, for example, 256 gradations with a dynamic range of "0 to 255".

<2. First embodiment> First embodiment of the present invention
An image data forming apparatus 300 according to the embodiment will be described. FIG. 3 is a block diagram illustrating a configuration of the image data forming apparatus 300 according to the first embodiment. The image data forming apparatus 300 includes an adder 11,
A memory address generation unit 12, a threshold memory 13, a binarization unit 14, an error detection unit 15, a distribution ratio memory 16, an error distribution unit 17, and an error memory 19 are provided.

The input image data SP for the target pixel P to be binarized is guided to the adder 11 when input to the image data forming apparatus 300. Then, the adder 11 adds the error EP that is generated from the pixels that have already been subjected to the binarization processing and is the accumulated value of the errors distributed to the target pixel P, thereby generating error addition data SP ′. The error addition data SP ′ is sent to the binarization unit 14 and the error detection unit 15.

On the other hand, the memory address generating section 12 receives the position information (fx, fy) of the target pixel P and inputs the position information (f, f) of the target pixel P in the repetition area AR as described above.
x ', fy'). More specifically, as shown in FIG. 10, the remainder obtained by dividing the position fx on the Fx axis by the size N of the repetition area AR is fx ′, and the position fy on the Fy axis is the size N of the repetition area AR. Let fy 'be the remainder after dividing by. The position information (fx ′, fy ′) of the target pixel P in the repeating area AR obtained in this way is sent to the threshold memory 13 and the distribution ratio memory 16.

The threshold memory 13 stores a threshold calculated for each recording pixel existing in the repetition area AR by a computer in advance. Then, the threshold value corresponding to each recording pixel in the repeating area AR stored in the threshold memory 13 is set so that the position of the recording pixel in the repeating area AR can be read as an address. Therefore, in the threshold memory 13, the position information (f) of the pixel of interest P sent from the memory address generator 12.
By using x ', fy') as the address, the threshold value TP set for the target pixel P can be read. The threshold value TP for the pixel of interest P obtained in this way is sent to the binarization unit 14.

The binarizing section 14 compares the error addition data SP 'for the target pixel P with the threshold value TP to generate binary data BP. Specifically, if “SP ′ ≧ TP”, “BP = 1” is set, and if “SP ′ <TP”, “BP” is set.
P = 0 ”. Then, the binary data BP is sent to the error detection unit 15 and the distribution ratio memory 16 and is output as an output from the image data forming apparatus 300 to the image recording apparatus 4 in the subsequent stage.
Sent to 00. In the image recording apparatus 400, if the binary data BP is "0", dot recording is not performed on the target pixel P, and if the binary data BP is "1", dot recording is performed on the target pixel P.

Then, the error detecting section 15 performs the binarizing section 1
In step 4, an error eP caused by the binarization of the image data performed on the target pixel P is derived. Specifically, the error eP
Is determined by "eP = SP'-255.BP". Therefore, if the binary data BP is "0", the error eP is often a positive number, and if the binary data BP is "1", the error eP is often a negative number. Then, the error detection unit 1
The error eP obtained in step 5 is sent to the error distribution unit 17.

The distribution ratio memory 16 inputs the position information (fx ', fy') obtained by the memory address generator 12 and the binary data BP as addresses. In this embodiment, as shown in the error memory 19 in FIG. 3 as an example, the error eP is distributed to the positions of twelve binarized unprocessed pixels P0 to P11. The distribution ratio memory 16 stores the distribution ratio W0 of the error eP with respect to the binarized unprocessed pixels P0 to P11 in each of the case where the binary data BP is "1" and the case where the binary data BP is "0". W
11 is previously derived and stored. Therefore, the position information (fx ', fy') for the target pixel P and the binary data BP
Is input, 12 is located in the vicinity of the target pixel P.
The distribution ratios W0 to W11 can be obtained for the respective binarized unprocessed pixels P0 to P11.

Here, the distribution ratios W0 to W11 for the unprocessed pixels P0 to P11 will be described. This distribution ratio W0
W11 is obtained from the threshold values T0 to T11 for binarization set for each of the unprocessed pixels P0 to P11. The threshold values T0 to T11 are the same as the threshold values stored in the threshold value memory 13 for the unprocessed pixels P0 to P11.

The distribution ratio WQ for an arbitrary pixel PQ (where Q is an integer of 0 to 11) among the unprocessed pixels P0 to P11 indicates that the binarization result for the target pixel P is "1". (Ie, "BP = 1")

[0036]

(Equation 1)

[0037] Here, Ti is a threshold value set for the i-th pixel among the unprocessed pixels P0 to P11. TQ is a threshold set for the pixel PQ.

When the binarization result of the target pixel P is "0" (that is, when "BP = 0"), the distribution ratio WQ is:

[0039]

(Equation 2)

Is obtained by In Equation 2, Tmax
Is the maximum value among the threshold values for each pixel set in the repeating area AR shown in FIG.

Then, as described above, the distribution ratios WQ as shown in Expressions 1 and 2 are previously derived and stored in the distribution ratio memory 16. Then, in the distribution ratio memory 16, the unprocessed pixel P0 obtained by using the position information (fx ', fy') for the target pixel P and the binary data BP as addresses
The distribution ratios W0 to W11 for .about.P11 are sent to the error distribution unit 17.

The error distribution unit 17 distributes each of the binarized unprocessed pixels P0 to P11 from the error eP obtained from the error detection unit 15 and the distribution ratios W0 to W11 obtained from the distribution ratio memory 16. The following errors e0 to e11 are derived. Specifically, any pixel PQ among the unprocessed pixels P0 to P11
(Where Q is an integer of 0 to 11), the error eQ distributed is

[0043]

(Equation 3)

Is obtained. The errors e0 to e obtained for each of the binarized unprocessed pixels P0 to P11
11 is sent from the error distribution unit 17 to the error memory 19.

In the error memory 19, the errors e 0 to e sent from the error distribution unit 17 for each of the binarized unprocessed pixels P 0 to P 11 located near the pixel of interest P.
Add 11 Therefore, when the pixel of interest P sequentially moves with the progress of the binarization processing, an error due to the binarization of the pixel of interest P is accumulated in each pixel of the error memory 19. Then, when these pixels become the target pixel P, the error EP, which is the accumulated value of the errors distributed by the previous processing, is sent to the adder 11.

The image data forming apparatus 30 of this embodiment
At 0, the binarization process for the pixel of interest P is performed as described above.

In the image data forming apparatus 300 having the above configuration, when the binarization result of the target pixel P is "1", the error eP is often a negative number as described above, and is a relatively small value. is there. In this case, when the distribution ratio based on the equation 1 is applied, the small error eP is distributed more to the pixels for which the large threshold is set.
As a result, a small (negative) error eP is distributed to the outer edge of the halftone dot region DR where dot recording is difficult to be performed. Therefore, when the pixel corresponding to the outer edge of the halftone dot region DR becomes the pixel of interest, a negative number or a small error EP is added to the input image data SP in the adder 11, and the error correction data SP 'is also Become smaller. And
In the binarization unit 14, although the threshold value is large, the error correction data SP 'is small, so that the binarization result often becomes "0", and dot recording is not performed. In other words, when the binarization result of the target pixel P is “1”, pixels that are difficult to form halftone dots by setting the threshold value are further difficult to form halftone dots by distributing the error eP. It is possible to prevent dots from being recorded in the area, and it is possible to maintain the shape of the halftone dots in a good condition.

When the binarization result of the target pixel P is "0", as described above, the error eP is often a positive number and a relatively large value. Then, in this case, when the distribution rate based on Equation 2 is applied, a large error eP
Is distributed more to pixels for which a small threshold is set. As a result, a large (positive number) error eP is distributed to the center of the halftone dot region DR where dot recording is easily performed. Therefore, when the pixel corresponding to the center of the halftone dot region DR becomes the target pixel, the adder 11 adds a positive or large error EP to the input image data SP, and the error correction data SP ′ Also increases.
In the binarization unit 14, the error correction data SP 'is large despite the small threshold value, so that the binarization result often becomes "1", and dot recording is easily performed. In other words, when the binarization result of the pixel of interest P is “0”, the pixels that are likely to form halftone dots by setting the threshold value are made easier to form halftone dots by distributing the error eP. It is possible to prevent a white portion that is not dot-recorded from occurring in the portion, and it is possible to maintain a good dot shape.

That is, in the image data forming apparatus of this embodiment, the binarization result of the target pixel P is "1".
In either case or in the case of "0", the distribution ratio of the error eP is determined so as to maintain the dot shape well.

The distribution ratio WQ for an arbitrary pixel PQ (where Q is an integer of 0 to 11) among the unprocessed pixels P0 to P11 is represented by the following formulas 1 and 2 and the unprocessed pixels P0 to P11. P
When the 11 threshold values T0 to T11 are sorted in ascending order, the pixel PQ
If the threshold TQ for is Rth,

[0051]

(Equation 4)

May be obtained as follows. However, in Equation 4, when the binarization result is “1”, “K> 1”, and
When the value is "0", "1>K>0". Even in this case, similarly to the above, when the error eP is a small value (mostly a negative number), more errors eP are distributed to pixels having a relatively large threshold value, and the error eP is a large value (mostly a positive number). In some cases, more errors eP are distributed to pixels having a relatively small threshold value, and the shape of the halftone dot is kept good.

Actually, in the image data forming apparatus 300 of this embodiment, the result of binarizing the image when the input image data is all “128” is image-recorded, as shown in FIG. Recorded image. As can be seen from the recorded image shown in FIG. 4, according to the image data forming apparatus 300 of this embodiment, each halftone dot is recorded in a state in which the shape of the dot is well maintained.

On the other hand, in order to make a comparison with FIG. 4, the distribution e is not set for each pixel of interest as in this embodiment, and the error eP is simply evenly applied to the unprocessed pixels P0 to P11. FIG. 5 shows a recorded image in the case of distribution. In the recorded image of FIG. 5, the halftone dot shape is disturbed, and phenomena such as the presence of dots separated from one halftone dot and the presence of a white dot in the halftone dot occur. On the other hand, in the recorded image of FIG. 4, the dot shape is well maintained as described above.

FIG. 6 shows a recorded image when the striped image in which fine lines are arranged in the horizontal direction is processed by the image data forming apparatus 300 shown in this embodiment. In this embodiment, by dividing the error eP based on the distribution ratio set for each pixel, the input image data and
Since the consistency with the image data after the binarization is maintained, the occurrence of moiré can be suppressed as shown in FIG.

<3. Second embodiment> Second embodiment of the present invention
An image data forming apparatus 300 according to the embodiment will be described. The image data forming apparatus described in the first embodiment has the effect of suppressing the occurrence of moire, but on the other hand, causes the edges of the image to be disturbed.
The second embodiment also improves such edge disturbance of an image.

FIG. 7 is a block diagram showing a configuration of an image data forming apparatus 300 according to the second embodiment. The image data forming apparatus 300 includes an adder 11, a memory address generating unit 12, a threshold memory 13, a binarizing unit 14, and an error detecting unit 1, as in the first embodiment.
5, a distribution ratio memory 16, an error distribution unit 17, and an error memory 19 are provided.

The difference between the image data forming apparatus 300 according to the second embodiment and the image data forming apparatus according to the first embodiment is the method of distributing errors in the error distribution unit 17. In addition, other components are described in the first section.
This is the same as that shown in the embodiment. The description of the same processing units as those described in the first embodiment will be omitted.

In this embodiment, the error distribution unit 17
Is the error eP caused by binarizing the target pixel P,
The distribution ratios W0 to W11 obtained for the unprocessed pixels P0 to P11 located in the vicinity of the pixel of interest P and the input image data S0 to W10 corresponding to the unprocessed pixels P0 to P11, respectively.
It distributes based on S11.

The distribution ratios W0 to W11 are read out by addressing the position information (fx ', fy') of the target pixel P in the repetition area AR and the binary data BP, as in the first embodiment. It is. The data stored in the distribution ratio memory 16 is the same as in the first embodiment.

In addition to the distribution ratios W0 to W11 and the error eP, input image data S0 to S11 and binary data BP corresponding to the unprocessed pixels P0 to P11 are also guided to the error distribution unit 17. Then, the error distribution unit 17 first normalizes all the input image data S0 to S11 of the unprocessed pixels P0 to P11. For example, assuming that image data obtained as a result of normalization is SQ 'for an arbitrary pixel PQ (where Q is an integer of 0 to 11) among the unprocessed pixels P0 to P11, "SQ'
= SQ / 255 ". Since the dynamic range of the input image data SQ is “0 to 255”, the normalized image data SQ ′ is “0 ≦ SQ ′ ≦ 1”.

The error distribution unit 17 derives the distribution ratio WQ 'based on the normalized image data SQ' and the distribution ratios W0 to W11 (WQ) obtained from the distribution ratio memory 16. That is, the binary data BP for the target pixel P is "1".
In the case of,

[0063]

(Equation 5)

A new distribution ratio WQ ′ is derived from the above, and when the binary data BP for the target pixel P is “0”,

[0065]

(Equation 6)

A new distribution ratio WQ ′ is derived from the above. According to Equation 5 or Equation 6, new distribution ratios W0 'to all the 12 unprocessed pixels P0 to P11 located near the pixel of interest P are obtained.
Guide W11 '.

Then, the error distribution unit 17 is provided with the error detection unit 1.
5, the errors e0 to e11 to be distributed to each of the binarized unprocessed pixels P0 to P11 are derived from the error eP obtained from No. 5 and the new distribution ratios W0 'to W11'. Specifically, an error eQ distributed to an arbitrary pixel PQ (where Q is an integer of 0 to 11) among the unprocessed pixels P0 to P11 is as follows:

[0068]

(Equation 7)

Is obtained. The errors e0 to e obtained for each of the binarized unprocessed pixels P0 to P11
11 is sent from the error distribution unit 17 to the error memory 19.

In the error memory 19, the errors e 0 to e sent from the error distribution unit 17 for each of the binarized unprocessed pixels P 0 to P 11 located near the pixel of interest P.
Add 11 Therefore, when the pixel of interest P sequentially moves with the progress of the binarization processing, an error due to the binarization of the pixel of interest P is accumulated in each pixel of the error memory 19. Then, when these pixels become the target pixel P, the error EP, which is the accumulated value of the errors distributed by the previous processing, is sent to the adder 11.

In the image data forming apparatus 300 according to the second embodiment, the binarization process for the target pixel P is performed as described above.

The image data forming apparatus 30 of this embodiment
At 0, when the binarization result of the target pixel P is "1", the error eP is often a negative number and a relatively small value, as described above. Then, in this case, when the distribution rate based on Equation 5 is applied, a small error eP
Are allocated to pixels for which a large threshold value has been set and to pixels for which input image data is small. As a result, small pixels of the input image data have smaller values.

On the other hand, when the binarization result of the target pixel P is "0", the error eP is a positive number in many cases as described above, and is a relatively large value. Then, in this case, when the distribution rate based on Equation 6 is applied, a large error eP
Are distributed more to pixels for which a small threshold value is set and to which input image data is large. As a result, a large pixel of the input image data has a larger value.

That is, in the case of this embodiment, the distribution ratios W0 to W11 obtained by the magnitude relation of the threshold value set for each pixel, and the input image data S
Since the error eP is distributed on the basis of 0 to S11, the shape of the halftone dot is maintained well and the contrast of the input image tends to be enhanced. Therefore, it is possible to perform processing that does not disturb the edge of the image while maintaining the dot shape.

In other words, in the image data forming apparatus according to the present embodiment, whether the binarization result of the target pixel P is “1” or “0”, Is maintained, and the distribution ratio of the error eP that does not disturb the edges of the image is determined.

FIG. 8 shows a recorded image obtained by actually processing a striped image in which fine lines are arranged in the horizontal direction in the image data forming apparatus 300 shown in the second embodiment. In this embodiment, by distributing the error eP based on the distribution ratio WQ and the input image data SQ, it is possible to keep the halftone dot shape well and to save the edges of the image well. Therefore, as shown in FIG. 8, generation of moiré can be suppressed, and fine lines are clearly reproduced.

<4. Third embodiment> Third embodiment of the present invention
An image data forming apparatus 300 according to the embodiment will be described. The third embodiment is the same as the image data forming apparatus 300 shown in FIG. 7 in the second embodiment, and also improves the edge disturbance of an image.

In the error distribution unit 17, distribution is performed based on the normalized image data SQ 'and distribution ratios W0 to W11 (WQ) obtained from the distribution ratio memory 16, as in the second embodiment. Leads to the rate WQ '. In the third embodiment, when the binary data BP for the target pixel P is “1”,

[0079]

(Equation 8)

A new distribution ratio WQ ′ is derived from the above, and when the binary data BP for the target pixel P is “0”,

[0081]

(Equation 9)

Then, a new distribution ratio WQ ′ is derived. From Equations 8 or 9, new distribution ratios W0 'to all the twelve unprocessed pixels P0 to P11 located near the pixel of interest P are obtained.
Guide W11 '. However, when the input image data S0 to S11 are all "0", "WQ '= WQ" is exceptionally set.

Then, the error distribution unit 17 is provided with the error detection unit 1
The errors e0 to e11 to be distributed to each of the binarized unprocessed pixels P0 to P11 are derived from Equation 7 from the error eP obtained from Eq. 5 and the new distribution ratios W0 'to W11'. And
The errors e0 to e11 obtained for the binarized unprocessed pixels P0 to P11 are sent from the error distribution unit 17 to the error memory 19.

In the image data forming apparatus 300 according to the third embodiment, the binarization process for the target pixel P is performed as described above.

The image data forming apparatus 30 of this embodiment
At 0, when the binarization result of the target pixel P is "1", the error eP is often a negative number and a relatively small value, as described above. Then, in this case, when the distribution rate based on Equation 8 is applied, a small error eP
Are allocated to pixels for which a large threshold value has been set and to pixels for which input image data is small. As a result, as in the second embodiment, a small pixel of the input image data has a smaller value.

On the other hand, when the binarization result of the pixel of interest P is "0", the error eP is a positive number in many cases as described above, and is a relatively large value. Then, in this case, when the distribution rate based on Equation 9 is applied, a large error eP
Are distributed more to pixels for which a small threshold value is set and to which input image data is large. As a result, as in the second embodiment, a large pixel of the input image data has a larger value.

That is, in the case of this embodiment, the second
In the same manner as in the embodiment, in order to distribute the error eP based on the distribution ratios W0 to W11 obtained by the magnitude relation of the threshold values set for each pixel and the input image data S0 to S11 of those pixels, The shape of the halftone dots is kept well and the contrast of the input image tends to be enhanced. Therefore, it is possible to perform processing that does not disturb the edge of the image while maintaining the dot shape.

FIG. 9 shows a recorded image obtained by actually processing a stripe image in which fine lines are arranged in the horizontal direction in the image data forming apparatus 300 shown in the third embodiment. Also in this embodiment, by distributing the error eP based on the distribution ratio WQ and the input image data SQ, the halftone dot shape can be kept good and the edges of the image can be well preserved. Therefore, as shown in FIG. 9, generation of moiré can be suppressed, and fine lines are clearly reproduced.

<5. Modification> In the image data forming apparatus 300 shown in the first to third embodiments, the binary data BP is input to the distribution ratio memory 16.
Instead of the binary data BP, a sign bit of the error eP may be used.

When each of the above embodiments is applied to processing of a color image, the same processing may be performed for each color component.

Further, FIGS. 3 and 7 show a block diagram in the case where the image data forming apparatus 300 is constituted by hardware. However, the image data forming apparatus 300 is realized by causing a computer to execute a predetermined processing program. It is also possible to realize. That is, processing such as signal conversion in each block shown in FIGS. 3 and 7 may be realized by a program.

[0092]

As described above, according to the first or third aspect of the present invention, the input image data of the target pixel to be binarized has already been subjected to the binarization processing. Error addition data is generated by adding the accumulated values of errors generated from the pixels, and a threshold value for forming a halftone dot corresponding to the position of the pixel of interest is compared with the error addition data to generate a binary value for the pixel of interest. Error is generated by the binarization process for the pixel of interest, and if the result of the binarization process for the pixel of interest is “1” as a logical value, it is included in a predetermined area. While a larger error is distributed to a plurality of pixels in descending order of the threshold value corresponding to each pixel, if the result of the binarization process for the target pixel is “0” as a logical value, Within the area of For a plurality of pixels to be processed, more errors are distributed in ascending order of the threshold value corresponding to each pixel, so that the halftone dot shape can be kept good, and the image quality such as the occurrence of moire in the recorded image can be reduced. Can be suppressed.

According to the second or fourth aspect of the present invention, when the result of the binarization process for the target pixel is "1" as a logical value, a plurality of pixels included in a predetermined area While a larger error is distributed to the pixels in the order of smaller image data for each pixel, if the result of the binarization processing for the pixel of interest is “0” as a logical value, a predetermined For a plurality of pixels included in the region, more errors are distributed in the order of larger image data for each pixel, so that the halftone dot shape can be kept good and the edges of the image are also good. Therefore, deterioration in image quality such as occurrence of moire in a recorded image can be favorably suppressed.

[Brief description of the drawings]

FIG. 1 is a schematic diagram of a configuration of an image processing system to which an embodiment of the present invention is applied;

FIG. 2 is a diagram showing an image composed of pixels in p rows and q columns.

FIG. 3 is a block diagram illustrating a configuration of an image data forming apparatus according to the first embodiment of the present invention.

FIG. 4 is a diagram showing a recorded image obtained by the image data forming apparatus according to the first embodiment of the present invention.

FIG. 5 is a diagram showing a recorded image when errors are evenly distributed.

FIG. 6 is a diagram illustrating a recorded image when a thin line image is processed in the image data forming apparatus according to the first embodiment of the present invention.

FIG. 7 is a block diagram illustrating a configuration of an image data forming apparatus according to a second embodiment of the present invention.

FIG. 8 is a diagram illustrating a recorded image when a thin line image is processed in the image data forming apparatus according to the second embodiment of the present invention.

FIG. 9 is a diagram illustrating a recorded image when a thin line image is processed in the image data forming apparatus according to the third embodiment of the present invention.

FIG. 10 is a diagram for explaining a halftone dot forming method.

FIG. 11 is a diagram showing a recorded image by the conventional halftone dot formation.

FIG. 12 is a diagram showing a recorded image when a thin line image is recorded by the conventional halftone dot formation.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Adder 12 Memory address generation part 13 Threshold memory 14 Binarization part 15 Error distribution part 16 Distribution rate memory 17 Error distribution part 19 Error memory 300 Image data forming device

Claims (4)

[Claims] 1. A method of performing binarization processing on multi-valued input image data input for each pixel to generate binarized image data, comprising: (a) a target pixel to be binarized Generating error addition data by adding a cumulative value of errors generated from pixels that have already been subjected to the binarization processing to the input image data of (b) corresponding to the position of the pixel of interest Performing a binarization process on the target pixel by comparing a threshold value for forming a halftone dot with the error addition data, and (c) an error generated by the binarization process on the target pixel. And (d) when the result of the binarization processing on the pixel of interest is “1” as a logical value, a plurality of pixels included in a predetermined area are In the descending order of the threshold value corresponding to the pixel, And (e) when the result of the binarization process for the pixel of interest is “0” as a logical value, a plurality of pixels included in the predetermined area And distributing the error more in ascending order of the threshold value corresponding to each pixel. 2. The method according to claim 1, wherein the step (d) includes: (d-1) when a result of the binarization processing on the target pixel is “1” as a logical value; For a plurality of pixels included in the predetermined area, further distributes the error more in order of image data for each pixel, the step (e), (e) -1) If the result of the binarization process for the target pixel is “0” as a logical value, the image data for each pixel is output for a plurality of pixels included in the predetermined region. A step of further distributing the error in ascending order of magnitude. 3. An apparatus for performing binarization processing on multivalued input image data input for each pixel to generate binarized image data, comprising: (a) a target pixel to be binarized; Adding means for generating error addition data by adding a cumulative value of errors generated from pixels that have already been subjected to the binarization processing to the input image data of (b) at the position of the pixel of interest Binarizing means for performing the binarizing process on the pixel of interest by comparing a threshold value for forming a corresponding halftone dot with the error-added data; and (c) the binarizing means in the binarizing means. Error detection means for detecting an error generated by the binarization process; and (d) when a result of the binarization process for the pixel of interest is “1” as a logical value, a plurality of pixels included in a predetermined area Pixels corresponding to each pixel In the descending order of the threshold value, the error distribution ratio is set to be large, and when the result of the binarization process for the target pixel is “0” as a logical value, a plurality of errors included in the predetermined region are included. A distribution ratio setting unit that sets the distribution ratio of the error to be large in the order of smaller threshold values corresponding to the pixels, and (e) a predetermined distribution ratio based on the distribution ratio obtained by the distribution ratio setting unit. And an error distribution unit that distributes the error to a plurality of pixels included in the region. 4. The apparatus according to claim 3, wherein the distribution ratio setting means is configured to: (d-1) when the result of the binarization processing on the pixel of interest is “1” as a logical value , For a plurality of pixels included in the predetermined region, the distribution ratio of the error is set to be larger in the order of smaller image data for each pixel, and the result of the binarization process for the pixel of interest is Is a logical value “0”, a plurality of pixels included in the predetermined area are
Means for setting the error distribution rate to be larger in the order of larger image data for each pixel.