KR100547102B1 - Adaptive error diffusion method based on the interval of black pixels - Google Patents

Abstract

The present invention relates to an error diffusion method in which a gray level error value between a pixel value of a continuous grayscale image and a thresholded binary pixel value is diffused to adjacent pixels according to a predetermined weight to perform binarization. When the error diffusion method is applied to the grayscale image, if the size of the pixel value of interest to be subjected to the halftone processing is included in the upper 90% or more of the gray scale range of the continuous grayscale image, the worm artifact and the In order to suppress the occurrence of start-up artifacts, the present invention relates to an adaptive error diffusion method based on black pixel spacing in which halftone processing is performed such that a distance from the nearest pixel halftone processed in black is kept constant.

According to the present invention, there is an advantage of effectively suppressing worm artifacts and start-up artifacts to obtain a high quality halftone image.

Description

Adaptive error diffusion method based on the interval of black pixels             

1 is an exemplary view showing a blue noise pattern in a halftone image;

2 is an exemplary view showing a worm artifact pattern in an error diffusion binarized halftone image;

3 is an exemplary view showing a binarized image in which start-up artifacts are generated when applied to a conventional error diffusion method.

4 is a block diagram showing a preferred embodiment of an adaptive error diffusion method based on black pixel spacing according to the present invention.

5a is an exemplary view showing an error diffusion mask proposed by Floyd and Steinberg,

5B is an exemplary diagram showing an error diffusion mask proposed by Stucky;

6 is an exemplary view showing a binarized image as a result of removing the start-up artifact by applying the adaptive error diffusion method based on black pixel spacing according to the present invention.

The present invention relates to a color halftone processing method in an image system, and more particularly to an error diffusion method, in which a distance from the closest pixel halftone processed in black in a continuous grayscale image is kept constant. The present invention relates to an adaptive error diffusion method based on black pixel spacing that effectively suppresses worm artifacts and start-up artifacts to obtain high quality halftone images.

Halftone processing is an image processing technique that makes a device appear to be close to the original image while reducing the quantization grayscale per pixel in a device having limited grayscale value reproduction characteristics. This technique can be used for various display devices such as printing presses, ink-jet printers, laser printers and liquid crystal displays (LCDs), and cathode ray tubes (CRTs). Widely used in image output apparatuses and the like.

로 인쇄 화상 데이터량을 감축할 수 있다.In particular, from the printing point of view, the halftone processing technique is a type of dot scattering and is the most commonly adopted image processing technique when printing image data in the form of a document together with dithering processing. Levels are expressed through the density of dots printed in the unit space, and are adopted as a printing method of most printed materials such as newspapers, books, promotional materials, and promotional flyers. Using this as an example, it is possible to convert the pixel of the image data represented by 8 bits to the 1-bit binary data most simply.

The amount of printed image data can be reduced.

In particular, relatively inexpensive printers are generally designed to print halftone images for ease of implementation and cost reduction, and in the field of printing a large quantity of prints, such as a print shop, halftone images are originally continuous. It is a long-standing issue to develop an appropriate conversion technique that can reduce the conversion time, which seems to be visually indistinguishable from the continuous tone image and at the same time has a significant effect on the printing speed.

The halftone processing method according to the prior art is the most well known ordered dither method (error diffusion method), dot diffusion method (dot diffusion method).

"Alignment" by Robert Ulichney and filed by "Digital Halftoning" and Chan, published by the Massachusetts Institute Technology press, obtained US Patent No. 5,140,432 Although these are described in detail in "Method and System for Reproducing Monochrome and Color Images Using Ordered Dither and Error Diffusion", the understanding of the present invention is understood. Briefly look at the halftone treatment method for the prior art in the present application as follows.

First, in contrast to the random dither method generating random thresholds, the alignment dither method has a simple thresholding process, that is, a simple thresholding process with a dither matrix that is a preset threshold array. In this method, the dither matrix is provided with a plurality of constituent elements to occupy physical space in the pixel space.

This dither matrix is mapped onto a continuous-tone image, and each pixel of the continuous grayscale image is compared pixel by pixel with each element of the corresponding dither matrix. If the pixel value of the continuous gradation image is larger than the element value (ie, the threshold value) of the corresponding dither matrix, the dot at the corresponding position of the halftone image is not printed, otherwise the dot is discarded. By printing, a natural halftone image is generated as a whole.

In this case, if the size of the space occupied by the dither matrix is smaller than the size of the continuous grayscale image, the dither matrix may be duplicated to fit the size of the entire continuous grayscale image so that halftone processing may be performed without being limited to the size of the image. have.

On the other hand, the error diffusion method, which is another method, is a threshold processing for assigning a gray scale value larger than the threshold value as a dot for printing based on a fixed threshold value and assigning the gray scale value to a dot for which no printing is performed. The continuous gradation image is converted and processed into a halftone image through a threshold process.

In this case, a lack of accuracy may occur due to the use of a fixed threshold value. Accordingly, in the error diffusion method, a difference between a target pixel value and a thresholded binary pixel value of a continuous grayscale image may occur. This problem is overcome by compensating for an error by diffusing the adjacent pixels.

In other words, the above-mentioned problem is solved by propagating an error between the binarization target pixel value and the thresholded binary pixel value of the continuous grayscale image being processed at present to adjacent pixels participating in the weight determination. This approach is even better when it comes to expressing continuous tones over a small area surrounded by some few dots.

In general, the error diffusion method has a relatively higher complexity than the alignment dither method, and thus requires more hardware components in implementation, but has an excellent gradation expression capability that can offset this in terms of image quality.

However, while the conventional error diffusion method has excellent gray scale representation capability showing the blue noise characteristic as shown in FIG. 1, the gray scale value is approximately 230 or more (that is, when referring to 256 gray scale images of 8 bits). In the case of belonging to a higher gradation value of 90% or more of the entire gradation range of the continuous gradation image), worm artifacts as shown in FIG. 2 have occurred.

Applicants of the present invention have also filed a Korean Patent Application No. 1998-031156, "Digestion Method for High Definition Printing," in order to remedy this problem, in particular, such worm artifacts are continuously generated throughout halftone images. There is a problem that serious visual deterioration in the binary halftone image because it is visually annoying.

In addition, the conventional error diffusion method has a disadvantage in that, in addition to the start-up artifacts, another kind of artifacts called start-up artifacts occurs.

3 is an exemplary view showing a binarized image in which start-up artifacts are generated when applied to a conventional error diffusion method.

As shown in FIG. 3, start-up artifacts are gradation errors of pixels belonging to a light gradation when moving from a region having a dark gradation to a region having a bright gradation when performing an error diffusion binarization process. Is small at first, and after processing a certain amount of pixels, pixels that are binarized to black do not appear.However, until the pixels are binarized to black, the white area is accumulated while the gray level error is accumulated. Deterioration occurs.

Accordingly, the present invention has been made to solve such a problem, and in the error diffusion method, the worm artifacts are maintained as the distance from the closest pixel half-tone processed in black in the continuous grayscale image is kept constant. It is an object of the present invention to provide an adaptive error diffusion method based on black pixel spacing, which effectively suppresses) and start-up artifacts to obtain high quality halftone images.

In order to achieve the above object, an adaptive error diffusion method based on black pixel spacing includes an adjacent pixel according to a predetermined weight based on a gray level error value between a pixel value of a continuous grayscale image and a thresholded binary pixel value. In the error diffusion method for diffraction and binarization, the size of the pixel value of interest to be subjected to halftone processing at the present time is the total gray level of the continuous grayscale image when the error diffusion method is applied to the continuous grayscale image. If it belongs to the gray level value of the upper 90% or more of the range, the halftone process is performed so that the distance between the closest pixel halftoned to black is kept constant to suppress the occurrence of worm artifacts and start-up artifacts. to be.

Hereinafter, a preferred embodiment of the adaptive error diffusion method based on black pixel spacing according to the present invention will be described with reference to FIG. 4.

4 is a block diagram showing a preferred embodiment of an adaptive error diffusion method based on black pixel spacing according to the present invention.

As shown in FIG. 4, a preferred embodiment of the adaptive error diffusion method based on black pixel spacing according to the present invention, the gray level error value between the pixel value of the binarized target image having continuous gray levels and the thresholded binary pixel value (gray In an error diffusion method of performing binarization by diffusing a level error value) to adjacent pixels according to a predetermined weight,

Designating a pixel of interest to be binarized in the binarization target image in raster scanning order (S10);

Calculating a corrected pixel value of interest by adding the gray level error value of the neighboring pixel adjacent to the pixel of interest to the pixel of interest (S20);

Determining whether the magnitude of the pixel value of interest is included in a preset artifact generation period through comparison of grayscale values (S30);

A first binarizing step (S40) of performing normal error diffusion binarization processing on the corrected pixel value of interest if the size of the pixel value of interest is not included in the upper grayscale value section; And

When the magnitude of the pixel value of interest is included in the artifact generation period, the binarization processing is performed on the pixel of interest using a first black pixel distance, which is a distance from the pixel that is binarized with the black color that exists on the shortest distance from the pixel of interest. It consists of a second binarization step (S50) to perform.

Here, the first binarization step (S40) may include determining whether the corrected pixel value of interest is greater than or equal to the halftone value (S41);

As a result of the determination in S41, if the corrected pixel value of interest is greater than or equal to the intermediate gray value, the pixel of interest is binarized to white, and then the new gray level error is subtracted from the corrected pixel value of interest. Assigning a value (S42); And

As a result of the determination in S41, if the corrected pixel value of interest is smaller than the intermediate gray level value, the pixel of interest pixel is binarized to black, and then the modified pixel value of interest is designated as a new gray level error value (S43). It is composed of

Here, the second binarization step (S50) may include determining whether the first black pixel separation distance is smaller than a preset forced separation distance (S51);

If the first black pixel separation distance is smaller than the forced separation distance as a result of the determination in operation S51, the pixel of interest is binarized to white, and the new gray level is subtracted from the corrected pixel value of correction. Specifying as an error value (S52); And

If the first black pixel separation distance is not smaller than the forced separation distance as a result of the determination in operation S51, after binarizing the pixel of interest to black, designating the corrected pixel value of interest as a new gray level error value. It consists of S53.

On the other hand, it is preferable that the artifact generation section is a gradation value section of at least 90% of the entire gradation range of the gradation image.

For example, when the number of bits per pixel is 8 bits, the intermediate gray value is 127, the maximum gray value is 255, and the worm artifact interval is preferably 230 or more.

The forced spacing distance is preferably a sum of a predetermined fixed spacing distance value and a randomly generated distance value within a range of ± 25% of the fixed spacing distance according to the number of bits per pixel.

With reference to the accompanying drawings, a process of carrying out the preferred embodiment of the adaptive error diffusion method based on the black pixel spacing according to the present invention configured as described above will be described in detail.

First, in step S10, the pixel of interest to be binarized in the binarization target image is designated in raster scanning order.

Subsequently, in step S20, the corrected pixel value of interest is calculated by adding the gray level error value of the peripheral pixel adjacent to the pixel of interest to the pixel of interest.

In operation S30, it is determined whether the magnitude of the pixel value of interest is included in a preset artifact generation interval by comparing grayscale values, wherein the artifact generation interval is a higher than 90% of the grayscale value interval of the entire grayscale range of the grayscale image. Set to. When the number of bits per pixel is 8 bits, the worm artifact interval is usually set to 230 or more, and the interval is an interval determined experimentally because the interval is determined by an experimental method rather than the interval determined by a special theoretical basis. It is well known that the fluctuations can be made around 220 ~ 230.

As a result of the determination in step S30, when the size of the pixel value of interest is not included in the upper grayscale value section, in the first binarization step S40, a normal error diffusion binarization process is performed on the corrected pixel value of interest.

As is well known, a representative method of known error diffusion methods is a method proposed by Floyd and Steidberg, and another representative method is Stucki.

FIG. 5A is an error diffusion mask proposed by Floyd and Steinberg, and FIG. 5B is an error diffusion mask proposed by Stucky.

For example, an error diffusion mask proposed by Floyd-Steinberg is shown in FIG. 5A, and the numbers described in each cell of the error diffusion mask are represented by In other words, the error diffusion method is the most basic premise that the error diffusion method should store the gray level error value of the previous line through the line memory.

의 주목 화소값

과 인접 화소들의 계조 오차값들

에 상관성 정도에 따라 각각 1/16, 5/16, 3/16, 7/16의 가중치를 적용하여 수정 주목 화소값

을 산출한다. 여기서,

는 수직 좌표(vertical coordinate)이며,

는 수평 좌표(horizontal coordinate)이다.As in Equation 1, the current coordinate position

Attention pixel value of

Tone Error Values of N and N

Attention pixel value modified by applying weights of 1/16, 5/16, 3/16, and 7/16, respectively, according to the degree of correlation

To calculate. here,

Is the vertical coordinate,

Is the horizontal coordinate.

을 산출한 후, 수학식 2와 같은 조건식에 의해 현 좌표 위치

에 대한 이치화 결과치

를 산출하고, 해당 위치에 대한 계조 오차값

을 산출한다.Corrected attention pixel value through Equation 1

After calculating the current position of the coordinates by the conditional expression

Binarization result for

To calculate the gradation error value

To calculate.

이 기설정된 고정 역치값(

, 화소당 비트수가 8비트인 256 계조 화상을 기준할 때, 127이 됨)보다 크거나 같으면, 이치화 결과치

로 백색(WHITE)을 출력한 후, 주목 화소값

에서 최대 계조값(화소당 비트수가 8비트인 256 계조 화상을 기준할 때, 255가 됨)을 감산하여 해당 위치에 대한 계조 오차값

으로 지정하고, 그렇지 않으면, 이치화 결과치

로 흑색(BLOCK)을 출력한 후, 주목 화소값

을 해당 위치에 대한 계조 오차값

으로 지정한다.In other words, the modified pixel value of interest

This preset fixed threshold value (

Is greater than or equal to 127 when referring to a 256-gradation image having 8 bits per pixel).

Pixel value of interest after outputting white (WHITE)

The gray level error value for the position is subtracted by subtracting the maximum gray value (which becomes 255 when referring to 256 gray scale images with 8 bits per pixel)

Otherwise, the binarization result

Pixel value of interest after outputting black (BLOCK)

Is the gradation error value for that position.

To be specified.

In order to summarize the above, the process of performing the first binarization step S40 will be described with reference to FIG. 3. First, in step S41, it is determined whether the corrected pixel value of interest is greater than or equal to the intermediate gradation value. .

As a result of the determination in the determining step S41, if the corrected pixel value of interest is greater than or equal to the intermediate gray value, in step S42, the pixel of interest is binarized to white, and then the maximum gray value is subtracted from the corrected pixel value of interest. Otherwise, the pixel of interest is binarized to black in step S43, and then the corrected pixel of interest is designated as a new tone error value.

On the other hand, when the size of the pixel of interest is included in the artifact generation period as a result of the determination in step S30, when the size of the pixel of interest is included in the artifact generation period in the second binarization step S50, the first black color is included in the artifact generation period. A binarization process is performed on the pixel of interest using the pixel separation distance.

The process of performing the second binarization step S50 will be described in more detail to help understand the present invention. First, in step S51, it is determined whether the first black pixel separation distance is smaller than the predetermined forced separation distance.

)로는 수학식 3과 같이, 화소당 비트수에 따라 기설정된 고정 이격 거리값(

)과 상기 고정 이격 거리값(

)의 ±25% 범위 내에서 랜덤 함수(random function)를 통해 발생시킨 거리값(

)을 합산한 값을 이용한다.At this time, the forced separation distance (

), As shown in Equation 3, the fixed distance distance value (

) And the fixed separation distance value (

Distance value generated by random function within ± 25% of

) Is used.

)은 입력 화상의 계조수에 따라 고정된 역치값이고, 통상, 특별한 이론적인 근거에 의해 정해진 값이라기 보다는 모의 실험을 통해 실험적인 방법으로 정하는 값이다.Where the fixed separation distance value (

) Is a fixed threshold value according to the number of gray levels of the input image, and is usually a value determined by an experimental method through simulation rather than a value determined by a special theoretical basis.

If the first black pixel separation distance is smaller than the forced separation distance as a result of the determination in step S51, in step S52, the pixel of interest is binarized to white and the maximum gray value is subtracted from the corrected pixel value of interest. If the first black pixel separation distance is not smaller than the forced separation distance, in step S53, the pixel of interest is binarized to black, and the correction is performed. The pixel value of interest is specified as a new gradation error value.

When all of these processes are completed, the process returns to step S10 to specify the next pixel of interest in raster scanning order, and repeats the same process until the halftone processing for all the pixels is completed.

6 is an exemplary view showing a binarized image as a result of removing the start-up artifact by applying the adaptive error diffusion method based on black pixel spacing according to the present invention.

According to the present invention, as can be seen through FIG. 6, it can be seen that the worm artifacts as shown in FIG. 2 and the start-up artifacts as shown in FIG. 3 are effectively suppressed.

)보다 가까울 때는 무조건 주목 화소를 백색으로 이치화하여 부정적인 상관 패턴(correlated patterm)인 워엄 아티팩트(worm artifact)의 발생을 억제하고 이 거리가 일정한 거리(

) 이상이 될 경우에는 무조건 주목 화소를 흑색으로 이치화하여 스타트업 아티팩트(start-up artifact)의 발생을 억제한다.The above-described core idea of the present invention is to remove the worm artifacts and the start-up artifacts by removing the worm artifacts and the start-up artifacts by keeping the separation distance from the pixel displaced to the black pixels located on the shortest distance from the target pixel at a predetermined forced separation distance. The pattern in gradation has a blue noise characteristic. In other words, from among the pixels which have already been binarized, the nearest pixel treated with black is searched to find a distance from the pixel of interest, and the distance is a constant distance (

), The pixel of interest is binarized unconditionally to suppress the occurrence of worm artifacts, which are correlated patterms.

), The pixel of interest is binarized unconditionally to black to suppress the occurrence of start-up artifacts.

Terminologies used herein are terms defined in consideration of functions in the present invention, which may vary according to the intention or customs of those skilled in the art, and the definitions should be based on the contents throughout the present application. will be.

In addition, since the present invention has been described through the preferred embodiment of the present invention, in view of the technical difficulty aspects of the present invention, those having ordinary skill in the art can easily be different from another embodiment of the present invention. Since modifications may be made, it is obvious that both the embodiments and modifications cited in the above description belong to the claims of the present invention.

As described above, in an error diffusion method in which gray level error values between a pixel value of a continuous grayscale image and a thresholded binary pixel value are diffused to adjacent pixels according to a predetermined weight to perform binarization. When the error diffusion method is applied to the continuous grayscale image, if the size of the pixel value of interest to be subjected to the halftone processing at this point belongs to the upper 90% or more of the grayscale range of the grayscale range of the continuous grayscale image, According to the present invention, in which halftone processing is performed such that the distance between the closest pixel half-toned in black is kept constant to suppress the occurrence of warm artifacts and start-up artifacts, worm artifacts and start-up There is an advantage in that high-quality halftone images can be obtained by effectively suppressing start-up artifacts.

Claims (6)

In an error diffusion method of performing binarization by diffusing a gray level error value between a pixel value of a binarization target image having a continuous gray level and a thresholded binary pixel value to adjacent pixels according to a predetermined weight, Designating a pixel of interest to be binarized in the binarization target image in raster scanning order; Calculating a corrected pixel value of interest by adding the gray level error value of a peripheral pixel adjacent to the pixel of interest to the pixel of interest; Determining whether the magnitude of the pixel value of interest is included in a preset artifact generation period by comparing grayscale values; If the magnitude of the pixel value of interest is not included in the upper grayscale value section, a normal error diffusion binarization process is performed on the correction pixel value of interest, and it is determined whether the pixel value of correction pixel is greater than or equal to the intermediate gray value. A first binarization step; And When the magnitude of the pixel value of interest is included in the artifact generation period, the binarization processing is performed on the pixel of interest using a first black pixel distance, which is a distance from the pixel that is binarized with the black color that exists on the shortest distance from the pixel of interest. Adaptive error diffusion method based on black pixel spacing, characterized in that it comprises a second binarization step of performing. The method of claim 1, wherein the first binarization step, If the corrected pixel value of interest is greater than or equal to the intermediate gray value, binarizing the pixel of interest to white, and subtracting the maximum gray value from the corrected pixel value of interest to designate a new gray level error value; And If the correction pixel value of interest is smaller than the intermediate grayscale value, the pixel of interest pixel is configured to be blacked, and then the pixel of interest pixel value is set to a new gray level error value. Adaptive Error Diffusion Method. The method of claim 1, wherein the second binarization step, Determining whether the first black pixel separation distance is smaller than a preset forced separation distance; If the first black pixel separation distance is smaller than the forced separation distance, binarizing the pixel of interest to white and subtracting the maximum gray value from the corrected pixel value of interest to specify a new gray level error value; And If the first black pixel separation distance is not smaller than the forced separation distance, the pixel of interest is binarized to black, and then the corrected pixel pixel value is designated as a new gray level error value. Adaptive error diffusion method based on. The adaptive error diffusion method of claim 1, wherein the artifact generation interval is a gradation value interval of at least 90% of the entire gradation range of the gradation image. The adaptive error based on black pixel separation according to claim 4, wherein when the number of bits per pixel is 8 bits, the intermediate gray value is 127, the maximum gray value is 255, and the worm artifact interval is 230 or more. Diffusion method. The method of claim 1, wherein the forced distance is a sum of a predetermined distance and a distance generated by a random function within a range of ± 25% of the fixed distance according to the number of bits per pixel. Adaptive error diffusion method based on black pixel spacing.

Images