JP2007074345A - Image processing apparatus, method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus which enhances the dispersability of dots to reduce the delay of dot generation and to reduce trailing phenomena. <P>SOLUTION: The image processing apparatus for multi-coding source image data constituted of M (M≥3) values into N(M>N≥2) values includes a data acquisition means 10 for acquiring source image data, a plurality of multi-coding means 20 for converting the source image data acquired by the data acquisition means 10 to K<SB>1</SB>, K<SB>2</SB>, K<SB>3</SB>, ... K<SB>n</SB>(N≥K<SB>1</SB>, K<SB>2</SB>, K<SB>3</SB>, ... K<SB>n</SB>≥2) values, an image synthesizing means 30 for synthesizing a plurality of pieces of image data multi-coded by the multi-coding means 20 to generate image data constituted of N values, a noticed pixel movement pattern storage means 21 wherein a plurality of patterns different by noticed pixel moving methods in an error diffusion method are stored, and a noticed pixel movement pattern selection means 22 for selecting a specific pattern from a plurality of noticed pixel movement patterns. The multi-coding means 20 convert the source image data to K<SB>1</SB>, K<SB>2</SB>, K<SB>3</SB>, ... K<SB>n</SB>values by the error diffusion method or a dither method. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image processing apparatus that multi-values original image data and performs image processing, and in particular, image processing that delays dot generation and reduces tailing for a plurality of original image data having the same information. The present invention relates to an apparatus, an image processing method, and an image processing program.

Conventionally, when printing a plate for creating a printed matter by a technique using a plate based on image data, it is difficult to be affected by the displacement of the plate, and a good printed matter can be obtained even if the plate is displaced. As a method for creating image data capable of printing a plate, there is an apparatus described in Patent Document 1.
This image processing apparatus converts multi-value image data corresponding to one color obtained by color development into a plurality of multi-value image data in image processing for processing input multi-value image data. A device that converts each of a plurality of multi-valued image data obtained by conversion into binary image data and synthesizes the obtained plurality of binary image data. However, in order to obtain a good printed matter, a contrivance is made to vary the distribution coefficient in the error diffusion method. By printing a plate based on the binary image data obtained by this process, even if the position of the plate is shifted when generating a printed material from the printed plate, the deviation is not noticeable and good printed material. Is what you get.
JP 2001-16437 A

However, in the above-described apparatus, the distribution coefficient in the error diffusion method is changed to make the positional deviation inconspicuous, and the dot generation delay and tailing phenomenon are not improved, and the image quality is low particularly in a gradation image. There was a problem.
In addition, since the process of calculating the logical sum of a plurality of binary image data is also discarding the density information of the original image data, degradation due to a decrease in gradation occurs, and dots are generated between the plurality of binary image data. When data to be recorded overlaps, processing for expanding data indicating dot recording around the overlapping position causes a problem that image quality deteriorates.

This invention has been made in view of the above-described circumstances, and when performing a plurality of image processing on the original image data and synthesizing the plurality of generated image data, the dispersibility of the dots is improved, An object of the present invention is to provide an image processing apparatus, an image processing method, and an image processing program that reduce dot generation delay and tailing phenomenon.
Another object of the present invention is to provide an image processing apparatus image processing method and an image processing program in which the gradation of an original image is not lost by adding a plurality of multi-value images.

[Mode 1] In order to achieve the above object, the image processing apparatus according to mode 1 uses N (M> N ≧ 2; N is a natural number) as original image data composed of M (M ≧ 3; M is a natural number) value. ) an image processing apparatus for multi-level to values, the data acquisition means for acquiring original image data, wherein K ₁ the original image data acquired by the data acquisition _{unit, K 2, K 3, ...} K n (n ≧ K ₁ , K ₂ , K ₃ ,... K _n ≧ 2; K ₁ , K ₂ , K ₃ ,... K _n are natural numbers) values, and error diffusion in the multi-value converting means Pixel-of-interest movement pattern storage means for storing a plurality of patterns with different pixel-of-interest movement methods, pixel-of-interest movement pattern selection means for selecting a specific pattern from the plurality of pixel-of-interest movement patterns, and pixel-of-interest movement pattern The multi-value converted by the multi-value conversion means using It synthesizes the image data, and having an image combining means for generating image data consisting of N value.
According to this configuration, by synthesizing multi-valued images generated by multi-value conversion means having different moving methods of the target pixel, dot generation delay of generated image data, tailing phenomenon reduction, and the like are achieved. And the dot dispersibility can be improved.

[Embodiment 2] Embodiment 2 relates to the image processing apparatus of Embodiment 1, wherein the multi-value conversion means converts the original image data into K ₁ , K ₂ , K ₃ ,... K _n values by an error diffusion method. An error diffusion matrix storage means for storing a plurality of error diffusion matrices having different distribution patterns in the error diffusion method; and an error diffusion matrix selection means for selecting a specific pattern from the plurality of error diffusion matrices. It is characterized by that.
According to this configuration, it is possible to improve the dot dispersibility of the generated image data by making the error diffusion matrices different in the error diffusion method.

[Mode 3] Mode 3 relates to the image processing apparatus according to mode 1, wherein the multi-value conversion means converts the original image data into K ₁ , K ₂ , K ₃ ,... K _n values by a dither method. And a dither matrix storage unit that stores a plurality of different dither matrices, and a dither matrix selection unit that selects a specific pattern from the plurality of dither matrices.
According to this configuration, by using the error diffusion method and the dither method together, it is possible to achieve both dot dispersibility and high-speed processing, and by generating different dither matrices in the dither method, the generated image The effect that the dispersibility of the dot of data can be improved is acquired.

[Embodiment 4] Embodiment 4 relates to the image processing apparatus of embodiment 1, wherein the image composition means synthesizes image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value quantization means. It is characterized by adding a plurality of image data.
According to this configuration, by adding a plurality of pieces of image data, it is possible to obtain N-value image data that does not lose gradation.

[Embodiment 5] Embodiment 5 relates to the image processing apparatus according to embodiment 4, wherein the image composition means arbitrarily selects image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value quantization means. Is converted by a coefficient L (L is a natural number), and a plurality of image data is added.
According to this configuration, when a plurality of images are combined, by converting the plurality of image data using coefficients, it is possible to adjust the gradation of the combined image and to weight the multi-value processing. An effect is obtained.

[Mode 6] Mode 6 relates to the image processing apparatus of mode 1, wherein the image synthesis means synthesizes image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value quantization means. It is characterized by taking a logical sum of binary image data.
According to this configuration, by taking the logical sum of the binary image data, it is possible to improve the dot generation delay, the tailing phenomenon, the dispersibility, and the like without increasing the gradation.

[Mode 7] Mode 7 relates to the image processing apparatus according to mode 6, wherein the image composition means arbitrarily selects K ₁ , K ₂ , K ₃ ,... K _n value image data obtained by the multi-value quantization means. The binary image data is converted into binary image data by the coefficient, and the logical sum of these binary image data is taken.
According to this configuration, when combining a plurality of images, the plurality of image data is converted into binary image data by an arbitrary coefficient, and then a logical sum of these binary image data is taken to obtain a gradation. The effect of improving dot generation delay, tailing phenomenon, dispersibility, and the like can be obtained without increasing.

[Embodiment 8] The image processing program of Embodiment 8 is an image that multi-values original image data composed of M (M ≧ 3; M is a natural number) values into N (M> N ≧ 2; N is a natural number) values. In the processing program, a data acquisition step for acquiring the original image data, and the original image data acquired by the data acquisition step are represented by K ₁ , K ₂ , K ₃ ,... K _n (N ≧ K ₁ , K ₂ , K ₃ , ... K _n ≥2; K ₁ , K ₂ , K ₃ , ... K _n are natural numbers) a plurality of multi-value conversion steps, and a plurality of different methods of moving the target pixel of the error diffusion method in the multi-value conversion step The pixel movement pattern storing step for storing the pattern of the pixel, the pixel movement pattern selection step for selecting a specific pattern from the plurality of pixel movement patterns of interest, and the multi-value conversion step using the pixel movement pattern for multi-value Is by synthesizing a plurality of image data, characterized in that it is a program for executing the image data image synthesis step of generating, realized as the processes in the computer consists of N values.
According to this configuration, the same effect as in the first aspect can be obtained.

[Mode 9] Mode 9 relates to the image processing program of mode 8, wherein the multi-value conversion step converts the original image data into K ₁ , K ₂ , K ₃ ,... K _n values by an error diffusion method. An error diffusion matrix storage step for storing a plurality of error diffusion matrices having different distribution patterns in the error diffusion method, and an error diffusion matrix selection step for selecting a specific pattern from the plurality of error diffusion matrices. It is characterized by that.
According to this configuration, the same effect as in the second mode can be obtained.

[Mode 10] Mode 10 relates to the image processing program of mode 8, and the multi-value conversion step converts the original image data into K ₁ , K ₂ , K ₃ ,... K _n values by a dither method. A dither matrix storing step for storing a plurality of different dither matrices, and a dither matrix selecting step for selecting a specific pattern from the plurality of dither matrices.
According to this configuration, the same effect as in the third aspect is obtained.

[Mode 11] Mode 11 relates to the image processing program of mode 8, wherein the image synthesis step synthesizes image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value conversion step. It is characterized by adding a plurality of image data.
According to this configuration, the same effect as in the fourth aspect can be obtained.

[Twelfth Embodiment] A twelfth embodiment relates to an image processing program of the eleventh embodiment, wherein the image synthesis step arbitrarily selects image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value conversion step. Is converted by a coefficient L (L is a natural number), and a plurality of image data is added.
According to this configuration, the same effect as in the fifth aspect can be obtained.

[Mode 13] Mode 13 relates to the image processing program of mode 8, wherein the image synthesis step synthesizes image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value conversion step. It is characterized by taking a logical sum of binary image data.
According to this configuration, the same effect as in the sixth aspect is obtained.

[Embodiment 14] Embodiment 14 relates to the image processing program of embodiment 13, wherein the image synthesizing step arbitrarily selects K ₁ , K ₂ , K ₃ ,... K _n value image data obtained by the multi-value conversion step. The binary image data is converted into binary image data by the coefficient, and the logical sum of these binary image data is taken.
According to this configuration, the same effect as in the seventh aspect can be obtained.

[Mode 15] An image processing method according to mode 15 is an image in which original image data composed of M (M ≧ 3; M is a natural number) value is multivalued to N (M> N ≧ 2; N is a natural number) value. In the processing method, a data acquisition step for acquiring the original image data, and the original image data acquired by the data acquisition step are represented by K ₁ , K ₂ , K ₃ ,... K _n (N ≧ K ₁ , K ₂ , K _{3 , ... K n ≧ 2; K} 1, K 2, K 3, ... K n is a plurality of the multi-level step, method of moving the target pixel error diffusion method in the multi-value conversion step of converting the natural number) value A pixel-of-interest movement pattern storage step that stores a plurality of different patterns, a pixel-of-interest movement pattern selection step that selects a specific pattern from the plurality of pixel-of-interest movement patterns, and the multi-value conversion step that uses the pixel-of-interest movement pattern Multi-valued by Synthesizing a plurality of image data, characterized by comprising an image synthesizing step of generating image data consisting of N value.
According to this configuration, the same effect as in the first aspect can be obtained.

[Mode 16] Mode 16 relates to the image processing method of mode 15, wherein the multi-value conversion step converts the original image data into K ₁ , K ₂ , K ₃ ,... K _n values by an error diffusion method. An error diffusion matrix storing step for storing a plurality of error diffusion matrices having different distribution patterns in the error diffusion method, and an error diffusion matrix selecting step for selecting a specific pattern from the plurality of error diffusion matrices. It is characterized by that.
According to this configuration, the same effect as in the second mode can be obtained.

[Mode 17] Mode 17 relates to the image processing method of mode 15, wherein the multi-value conversion step converts the original image data into K ₁ , K ₂ , K ₃ ,... K _n values by a dither method. A dither matrix storing step for storing a plurality of different dither matrices, and a dither matrix selecting step for selecting a specific pattern from the plurality of dither matrices.
According to this configuration, the same effect as in the third aspect is obtained.

[Mode 18] Mode 18 relates to the image processing method of mode 15, wherein the image synthesis step synthesizes image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value conversion step. It is characterized by adding a plurality of image data.
According to this configuration, the same effect as in the fourth aspect can be obtained.

[Embodiment 19] Embodiment 19 relates to the image processing method of embodiment 18, wherein the image synthesis step arbitrarily selects image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value conversion step. Is converted by a coefficient L (L is a natural number), and a plurality of image data is added.
According to this configuration, the same effect as in the fifth aspect can be obtained.

[Mode 20] Mode 20 relates to the image processing method of mode 15, wherein the image synthesis step synthesizes image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value conversion step. It is characterized by taking a logical sum of binary image data.
According to this configuration, the same effect as in the sixth aspect is obtained.

[Mode 21] Mode 21 relates to the image processing method according to mode 20, wherein the image synthesis step arbitrarily selects image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value conversion step. The binary image data is converted into binary image data by the coefficient, and the logical sum of these binary image data is taken.
According to this configuration, the same effect as in the seventh aspect can be obtained.

Hereinafter, an image processing apparatus according to an embodiment of the present invention will be described in detail with reference to the drawings. In the following description, M, N, K ₁ , K ₂ , K ₃ ,... K _n are natural numbers.
First, an image processing apparatus according to an embodiment of the present invention will be described with reference to FIGS.
FIG. 1 is a block diagram showing a configuration of an image processing apparatus according to an embodiment of the present invention.

The image processing apparatus 1 includes a data acquisition unit 10, a multi-value conversion unit 20, an image synthesis unit 30, a target pixel movement pattern storage unit 21, and a target pixel movement pattern selection unit 22.
The data acquisition unit 10 acquires original image data. The original image data is M (M ≧ 3) value image data.

The multi-value conversion means 20 converts the M-value original image data acquired by the data acquisition means 10 into a plurality of multi-value image data.
The target pixel movement pattern storage unit 21 stores a plurality of patterns, for example, pattern 1, pattern 2,..., Pattern n, which have different target pixel movement methods in the error diffusion method.
The target pixel movement pattern selection unit 22 selects a specific pattern from the plurality of target pixel movement patterns stored in the target pixel movement pattern storage unit 21.

The image synthesizing unit 30 synthesizes the multi-valued image data obtained by the multi-leveling unit 20, and synthesizes the multi-valued image data by adding or logically adding a plurality of image data. At the time of image composition, it is preferable to take a logical sum of binary image data or to take a logical sum after converting multi-valued image data into binary image data using a coefficient. In the present embodiment, multi-value original image data K ₁ , K ₂ , K ₃ ,... K _n are converted into N-valued data by this synthesis (M> N ≧ K ₁ , K ₂ , K ₃ ,... K). _n ≧ 2).

FIG. 2 is a block diagram showing the configuration of the multi-value quantization means.
The multi-value conversion means 20 converts the original image data into multi-value image data, an error diffusion matrix storage means 23, an error diffusion matrix selection means 24, a dither matrix storage means 25, and a dither matrix selection means 26. With.
The error diffusion matrix storage means 23 stores a plurality of error diffusion matrices having different distribution patterns in the error diffusion method, for example, matrix 1, matrix 2,.

The error diffusion matrix selection unit 24 selects a specific pattern from a plurality of error diffusion matrices stored in the error diffusion matrix storage unit.
The dither matrix storage means 25 stores a plurality of different dither matrices, for example, matrix 1, matrix 2,.
The dither matrix selection unit 26 selects a specific pattern from a plurality of dither matrices stored in the dither matrix storage unit 25.

Then, the multi-value conversion means 20 performs multi-value conversion using the specific pattern selected by the error diffusion matrix selection means 24 and the dither matrix selection means 26 as necessary. In the multi-value conversion method, the original image data of M value is converted to K ₁ , K ₂ , K ₃ ,... K _n (M> K ₁ , K ₂ , K ₃ ,... K _n ≧ 2) values by an error diffusion method. Convert and do.

Next, the processing operation of the image processing apparatus according to the embodiment of the present invention will be described.
FIG. 3 is a flowchart showing the processing operation of the image processing apparatus according to the embodiment of the present invention.
First, in step S31, the data acquisition unit 10 acquires M-value original image data.
Next, in step S32, the multi-value quantization means 20 compares the input luminance value of the target pixel of the original image data acquired by the data acquisition means 10 with the multi-value threshold, and multi-values the target pixel (K ₁ , K ₂ , K ₃ ,... K _n value).

Next, in step S <b> 33, the multi-value quantization unit 20 sets the difference between the luminance value after the multi-value processing of the target pixel and the input luminance value as an error, and diffuses the error around the target pixel.
Next, in step S34, it is determined whether or not error diffusion has been completed for all target pixels. If not, the target pixel is moved to the next target pixel according to the target pixel movement pattern, and steps S32 to S32 are performed. Similar to S33, multilevel processing and error diffusion processing are performed until error diffusion of all target pixels is completed.

If it is determined in step S34 that the target pixel has reached the end of the processing target image, the generated multivalued image data is synthesized by the image synthesis means 30 in step S35.
Next, in step S36, it is determined whether or not all the image data have been combined. If there is remaining image data as a result of the determination, the process returns to step S31 to acquire the original image of M value, and step S32 In the same manner as in step S35, the multilevel processing, error diffusion processing, and image synthesis are repeated.
When all image data are combined in step S36, an N-value image is generated in step S37 and the process ends.

Next, an image composition method by the image processing apparatus according to the embodiment of the present invention will be described.
FIG. 4 is a schematic diagram showing a method for synthesizing images. (A) is an original image, (b) is an attention pixel movement pattern, (c) is an image after binarization, and (d) is a diagram showing an image after composition.
First, as shown in FIG. 4A, the original image is represented by a diagram indicated by “A”. This image is an image having M (M ≧ 3) values. This M-value original image is made into three binarized images with different pixel movement patterns of interest.

As shown in FIG. 4B, the M-value original image is binarized by storing the error diffusion in one pixel as a target moving pixel moving pattern. Here, three patterns of attention pixel movement patterns are used as indicated by b1, b2, and b3 in FIG. That is, in the pattern b1, after moving from the left to the right in the uppermost stage (1), it moves to the lower stage, moves from the left to the right in (2), and repeats to (5) sequentially. In the pattern b2, after moving from the top to the bottom at (1) at the right end, it moves to the left at (2), and from top to bottom at (2), and repeats up to (6) sequentially. In the pattern b3, after moving from the right to the left in the lowermost stage (1), moving up, moving from the right to the left in (2), the process is repeated up to (5). Each image binarized in this way (K ₁ = 2, K ₂ = 2 and K ₃ = 2) is as shown in FIG.

In FIG. 4C, the binarized images for the patterns b1, b2, and b3 in FIG. 4B are as c1, c2, and c3, respectively. These binarized images are represented by one of 0 (white) and 1 (black), but since the target pixel movement pattern is different, there are three types of images.
Then, when these images shown in FIG. 4C are added and combined, an image of N (N = 4) values as shown in FIG. 4D is obtained. The synthesized image is represented as 0 (white), 1 (light ash), 2 (dark ash), 3 (black).
As described above, according to the present embodiment, M ≧ 3, K ₁ = 2, K ₂ = 2 and K ₃ = 2 and the original image data of M value is binarized with three types of patterns, and then combined. In this example, four-value image data is created.

[First embodiment] <2 value + 2 value → 3 value>
Next, an image processing apparatus according to the first embodiment of the present invention will be described with reference to FIGS. In this embodiment, an M-value original image is multi-valued into two binary images, and these images are combined to create ternary image data. Note that the ternarization processing by the error diffusion method of the present embodiment is multi-valued to (binary + binary) by the pixel-of-interest movement pattern selection means 22 with the pixel-of-interest movement direction as upper left → lower right and lower right → upper left, respectively. Then, the case where these images are combined and ternarized is shown.

FIG. 5 is a diagram showing an error diffusion matrix in the image processing apparatus according to the first embodiment of the present invention. In the error diffusion matrix in the present embodiment, as shown in the figure, the ratio of the distribution ratio is viewed with the individual pixel as the target pixel in the horizontal direction from the upper left. In the figure, the distribution rate of the pixel of interest is indicated by the fraction at the end of the arrow. The ratio is, for example, that the rightward portion adjacent to the target pixel is 7/48 right next to the pixel of interest and 5/48 next to it, and the lower portion adjacent to the target pixel is right below. The error diffusion is 7/48, and below that is 5/48. In addition, as for the portion in the oblique direction from the target pixel, the immediately oblique portion is 5/48, and the oblique portion is 1/48.
As described above, when the original image is converted into an image having a smaller number of gradations than the original gradation, the error diffusion method is adopted. This error diffusion method is a method for converting each image in the converted image. In this method, an error between a tone value and a gradation value in an original image is distributed to surrounding pixels.

Hereinafter, the basic principle of the error diffusion method will be described. In the following description, a case where a 256-tone halftone image is binarized and converted to 0 or 255 will be exemplified.
First, for a target pixel in the original image as shown in FIG. 5, the density value is converted to either 0 or 255 by comparing it with a predetermined threshold value. In this example, the pixel of interest starts with the pixel in the upper left corner and moves in the horizontal direction, which is the main scanning direction, and moves along the vertical direction, which is the sub-scanning direction, when processing for one line is completed in the main scanning direction. And move to the next line. Here, the threshold is assumed to be 128, which is an intermediate value of 256 (gradation).

  Initially, if the density value of the pixel in the upper left corner of the original image is 129, comparing this with the threshold value 128, 129> 128 and (density value of the target pixel)> (threshold value), and the density value of the target pixel is Is converted to 255. At this time, since an error (difference value) of 255-129 = 126 is generated by converting the target pixel from 129 to 255, this error is distributed to surrounding pixels based on the error diffusion processing table shown in FIG. Similarly, binarization is performed on the second pixel, in this example, the pixel on the right of the upper left corner pixel, the third pixel,... This error diffusion processing table is an example in which preset diffusion coefficients are shown in a matrix format, and a value obtained by multiplying a difference value in the target pixel by a diffusion coefficient is distributed to surrounding pixels.

  As shown in FIG. 5, the swing range when a 5 × 7 matrix is used for the error diffusion processing table is 5 × 7. In the error diffusion method, the larger the swinging range, the better the image quality of the photographic image with the high frequency components remaining in the converted image. In addition, before and after the error diffusion processing, the sum of diffusion coefficients in this error diffusion processing table is set to 1 so that the image density is maintained.

Here, for example, in the pixel on the right side of the target pixel, 126 (difference value of the target pixel) × 7/48 (diffusion coefficient in the pixel) −20 (the decimal part is rounded off or rounded down) is the distribution value. (However, the sign of the distribution value is opposite to the difference value of the target pixel in order to preserve the density of the entire image, and −20 is added to the pixel on the right side of the target pixel). Based on this error diffusion processing table, values assigned to surrounding pixels are calculated in the same manner.
In this way, when error diffusion for the first pixel of interest is completed, error diffusion is performed using the pixel on the right side of the pixel of interest as a new pixel of interest, and all pixels are sequentially shifted while the pixels of interest are sequentially shifted. By performing error diffusion on the original image, the original image can be converted into an image having a smaller number of gradations than the original gradation.

Next, with reference to FIG. 6, the effect of the ternarization processing by the error diffusion method in the image processing apparatus according to the first embodiment of the present invention will be described.
FIG. 6 is a diagram showing an image state in which the processing result by the error diffusion method according to the present invention is compared with a conventional example. (A) is an original image, (b) is an image and a partially enlarged view of a processing result according to the present invention, and (c) is a view showing an image and a partially enlarged view of a conventional example.

  According to the ternarization processing according to the present invention, the original image which is the M-value image data shown in FIG. 10A is generated at the left end as shown in the enlarged view of FIG. The delay and dot tailing are offset and the dispersibility is improved. In the halftone part, the dispersibility of different dot types is improved. At the right end, the dot generation delay and the dot tailing are improved. Is offset and dispersibility is improved.

The effect by (b) of the figure is remarkable when compared with the conventional example of (c) of the figure. That is, in the left end portion of FIG. 5C, the dot generation is delayed due to the influence of the diffusion direction of the error diffusion matrix, and the dot generation occurs so as to flow in the lower right direction. It can be seen that the dispersibility of the seeds has deteriorated, and in the right end portion, dots have flowed in the lower right direction due to the influence of the diffusion direction of the error diffusion matrix.
As described above, according to the present embodiment, dot generation delay and dot tailing are canceled out, and an effect of improving dispersibility can be obtained.

[Second Embodiment] <Binary + Binary → Ternary>
Next, with reference to FIG. 7, the ternarization processing by the error diffusion method in the image processing apparatus according to the second embodiment of the present invention will be described.
7A and 7B are diagrams showing ternarization processing by the error diffusion method in the image processing apparatus according to the second embodiment of the present invention, where FIG. 7A shows an error diffusion matrix, and FIG. 7B shows processing by the present embodiment. It is a figure which shows a result. Note that the ternarization processing by the error diffusion method of the present embodiment multi-values the target pixel movement direction from the upper left to the lower right and the lower right to the upper left, respectively, and then synthesizes these images. The case of ternarization is shown.

  In the error diffusion matrix shown in FIG. 7A, the ratio of the distribution ratio is seen from the upper left of the figure as the pixel of interest in the horizontal direction. The pixel of interest is indicated by arrows in three directions toward the right side, the lower side, and the lower right side of the drawing. In the ternary error diffusion matrix in this embodiment, the error ratio of the right and lower pixels of the target pixel is 3/8, and the error ratio of the lower right pixel is 2/8. Their sum is one.

Then, using the error diffusion matrix described above, a binarized image is added by two patterns of a pattern in which the target pixel movement direction is upper left → lower right and a target pixel movement pattern is lower right → upper left pattern to obtain 3 A valuated image was synthesized.
As a result, as shown in FIG. 7B, in the left end portion, the dot generation delay and the dot tailing are offset to improve the dispersibility, and in the halftone portion, the dispersion of different dot types is improved. In the right end portion, dot generation delay and dot tailing are offset, and dispersibility is improved.

[Third embodiment] <2 values + 2 values + dither values 2 → 4 values>
Next, with reference to FIG. 8, description will be given of quaternarization processing by the error diffusion method and dither method in the image processing apparatus according to the third embodiment of the present invention.
In this embodiment, a quaternarization method using an error diffusion method (binary + binary) and a dither method (binary) is shown. That is, error diffusion (binary value + binary value) + dither binary value = quaternarization is shown. Also in the present embodiment, a binarized image is used in which the target pixel movement direction is an upper left → lower right pattern and the target pixel movement pattern is a lower right → upper left pattern.

FIG. 8A shows an example of a dither matrix, and FIG. 8B shows a processing result according to this embodiment.
The dither matrix shown in FIG. 8A is obtained by dividing the original image data 0 to 255 into 16 parts, each having 16 threshold values as shown in the figure.
The dither method is a method of performing halftone processing according to the magnitude relationship between the gradation value of image data and a threshold value of a predetermined dither matrix. Since the image data is larger in size than the dither matrix, in the dither method, the halftone process is usually performed by associating the dither matrix with each pixel in a predetermined arrangement. In general, the dither matrix is set so as to ensure the dispersibility of dots within a single matrix. Therefore, when the dither matrix is arranged independently for each page, there may be a case where sufficient dot dispersibility cannot be secured at the boundary portion between the two. On the other hand, in the image processing apparatus according to the present embodiment, the dither matrix is allowed to be arranged across the boundary, so that the dot dispersibility can be sufficiently secured even at the boundary portion, and high-quality image processing is performed. Can be realized.

  As a result, as shown in FIG. 8B, in the left end portion, the dot generation delay and the dot tailing are offset to improve the dispersibility, and in the halftone portion, the dispersion of different dot types is improved. In the right end portion, dot generation delay and dot tailing are offset, and dispersibility is improved. That is, this method also cancels the dot generation delay, improves the dispersibility, and improves the dispersibility as a whole.

[Fourth embodiment] <2 values × 2 + 2 values × 2 + dither values 2 → 6 values>
Next, with reference to FIG. 9, description will be made on the hexarization processing by the error diffusion method and the dither method in the image processing apparatus according to the fourth embodiment of the present invention.
In the present embodiment, a hexarization method using an error diffusion method (binary × 2 + binary × 2) and a dither method (binary) is shown. That is, error diffusion (2 times of binary + 2 times of binary) + 1 of dither binary = 6 values. Also in the present embodiment, a binarized image is used in which the target pixel movement direction is an upper left → lower right pattern and the target pixel movement pattern is a lower right → upper left pattern.

FIG. 9A shows an example of a dither matrix, and FIG. 9B shows a processing result according to the present embodiment.
The dither matrix shown in FIG. 9A is the same as the dither matrix in the third embodiment described above, and is obtained by dividing the original image data 0 to 255 into 16 thresholds each having 16 thresholds as shown in the figure. Have a value. In the image processing apparatus according to the present embodiment, the dither matrix is allowed to be arranged across the boundary, so that the dot dispersibility can be sufficiently ensured even at the boundary portion, and high-quality image processing can be realized. it can.

  As a result, as shown in FIG. 9B, in the left end portion, the dot generation delay and the dot tailing are offset to improve the dispersibility, and in the halftone portion, the dispersion of different dot types is improved. In the right end portion, dot generation delay and dot tailing are offset, and dispersibility is improved. According to the present embodiment, since the synthesized image is a 6-value image, it can be seen that the effect of the error diffusion process is stronger than that of the third embodiment.

As described above, according to the present invention, in any of the embodiments, the dot generation delay is offset and the dispersibility is improved as compared with the conventional example, and the dispersibility of different dot types is improved in the halftone portion. In addition, dot tailing is canceled out, and dispersibility is improved.
In the above embodiment, the case where the control program stored in advance in the ROM is executed when executing the processing shown in the flowchart of FIG. 3 is described. However, the present invention is not limited to this, and the program showing these procedures is also described. The program may be incorporated into a RAM and executed from a storage medium on which is recorded. Alternatively, the program may be acquired from a network.

  Here, the storage medium is a semiconductor storage medium such as RAM or ROM, a magnetic storage type storage medium such as FD or HD, an optical reading type storage medium such as CD, CDV, LD, or DVD, or a magnetic storage type such as MO. / Optical reading type storage media, including any storage media that can be read by a computer regardless of electronic, magnetic, optical, or other reading methods.

As mentioned above, although embodiment of this invention was described in detail, this invention is not limited to these Examples, A various change is possible without deviating from the meaning of this invention.
For example, in the above-described embodiments, multi-value processing up to 6-values has been described. However, the present invention is not limited to this, and more multi-values can be performed if necessary.
In the above-described embodiments, an example of the error diffusion matrix and the dither matrix is shown. However, the present invention is not limited to this, and other error diffusion matrices and dither matrices can be used.

1 is a block diagram illustrating a configuration of an image processing apparatus according to an embodiment of the present invention. It is a block diagram which shows the structure of a multi-value conversion means. 5 is a flowchart showing an image processing operation in the image processing apparatus according to the embodiment of the present invention. It is a figure which shows the composition method of the image in the image processing apparatus which concerns on embodiment of this invention. (A) is an original image, (b) is a movement pattern of a target pixel, (c) is an image after binarization, and (d) is an image after composition. It is a figure which shows the error diffusion matrix which concerns on 1st Example of this invention. (A) which concerns on 1st Example of this invention is a figure which shows the original image, (b) Processing result, (c) The processing result by a prior art example. It is a figure which shows (a) error diffusion matrix which concerns on 2nd Example of this invention, and (b) process result. It is a figure which shows the (a) dither matrix and (b) process result which concern on 3rd Example of this invention. It is a figure which shows the (a) dither matrix and (b) process result which concern on 4th Example of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Image processing apparatus, 10 ... Data acquisition apparatus, 20 ... Multi-value conversion means, 21 ... Interesting pixel movement pattern, 22 ... Interesting pixel movement pattern selection means, 23 ... Error diffusion matrix storage means, 24 ... Error diffusion matrix selection means 25 ... Dither matrix selection means, 26 ... Dither matrix selection means, 30 ... Image composition means

Claims (9)

In an image processing apparatus that multi-values original image data composed of M (M ≧ 3; M is a natural number) values into N (M> N ≧ 2; N is a natural number) values,
Data acquisition means for acquiring the original image data;
The original image data acquired by the data acquisition means is K ₁ , K ₂ , K ₃ ,... K _n (N ≧ K ₁ , K ₂ , K ₃ ,... K _n ≧ 2; K ₁ , K ₂ , K ₃ , ... K _n is a natural number) a plurality of multi-value conversion means for converting to a value;
A pixel-of-interest movement pattern storage unit that stores a plurality of patterns in which the pixel-of-interest movement method of the error diffusion method in the multi-value conversion unit is different;
Pixel-of-interest movement pattern selection means for selecting a specific pattern from the plurality of pixel-of-interest movement patterns;
Image synthesizing means for synthesizing a plurality of image data multi-valued by the multi-value quantization means using the pixel movement pattern of interest and generating image data composed of N values;
An image processing apparatus comprising: The multi-value conversion means converts the original image data into K ₁ , K ₂ , K ₃ ,... K _n values by an error diffusion method, and a plurality of error diffusions having different distribution patterns in the error diffusion method. The image processing apparatus according to claim 1, further comprising: an error diffusion matrix storage unit that stores a matrix; and an error diffusion matrix selection unit that selects a specific pattern from the plurality of error diffusion matrices. The multi-value conversion means converts the original image data into K ₁ , K ₂ , K ₃ ,... K _n values by a dither method, and a dither matrix storage means for storing a plurality of different dither matrices. The image processing apparatus according to claim 1, further comprising: a dither matrix selecting unit that selects a specific pattern from the plurality of dither matrices. The image synthesizing means synthesizes image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value quantization means, and adds a plurality of image data. The image processing apparatus according to claim 1. The image synthesizing means converts the image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value converting means with an arbitrary coefficient L (L is a natural number), and converts a plurality of image data. The image processing apparatus according to claim 4, wherein addition is performed. The image synthesizing means synthesizes image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value conversion means, and takes a logical sum of the binary image data. The image processing apparatus according to claim 1. The image synthesizing means converts the image data of K ₁ , K ₂ , K ₃ ,... K _n values obtained by the multi-value conversion means into binary image data with an arbitrary coefficient, and then converts these binary values. 7. The image processing apparatus according to claim 6, wherein the logical sum of the image data is taken. In an image processing program that multi-values original image data composed of M (M ≧ 3; M is a natural number) values into N (M> N ≧ 2; N is a natural number) values,
A data acquisition step of acquiring the original image data;
The original image data acquired in the data acquisition step is represented by K ₁ , K ₂ , K ₃ ,... K _n (N ≧ K ₁ , K ₂ , K ₃ ,... K _n ≧ 2; K ₁ , K ₂ , K ₃ , ... K _n is a plurality of multi-level step for converting natural number) value,
A pixel-of-interest movement pattern storage step for storing a plurality of patterns in which the pixel-of-interest movement method of the error diffusion method in the multi-value conversion step is different;
A pixel-of-interest movement pattern selection step of selecting a specific pattern from the plurality of pixel-of-interest movement patterns, and a plurality of image data multi-valued by the multi-value conversion step using the pixel-of-interest movement pattern, An image composition step for generating image data composed of values;
An image processing program for causing a computer to execute processing realized as: In an image processing method for converting original image data composed of M (M ≧ 3; M is a natural number) values into N (M> N ≧ 2; N is a natural number) values,
A data acquisition step of acquiring the original image data;
The original image data acquired in the data acquisition step is represented by K ₁ , K ₂ , K ₃ ,... K _n (N ≧ K ₁ , K ₂ , K ₃ ,... K _n ≧ 2; K ₁ , K ₂ , K ₃ , ... K _n is a natural number) a plurality of multi-value conversion steps to be converted into values,
A pixel-of-interest movement pattern storage step for storing a plurality of patterns in which the pixel-of-interest movement method of the error diffusion method in the multi-value conversion step is different;
An attention pixel movement pattern selection step of selecting a specific pattern from the plurality of attention pixel movement patterns;
An image synthesis step of synthesizing a plurality of image data multi-valued by the multi-value quantization step using the target pixel movement pattern, and generating image data composed of N values;
An image processing method comprising:

Images