KR930007979B1 - Middle tone picture display method - Google Patents

Abstract

The edge of a letter is emphasized to increase decipherability and the image data is transmitted at high speed by a dithering method which uses wave dither matrix. The method comprises the steps of: (A) scanning a manuscript by pixel unit and storing the scanned data on a buffer; (B) deciding the local brightness difference of the current scan data; (C) discriminating edge by inspecting that the brightness difference is larger than local inverse value; (D) converting half tone to white when the brightness difference is larger than local inverse value and the current pixel is brighter than local brightness and otherwise converting to black; and (E) discriminating that current pixel is not edge when local brightness difference is less than local inverse value and coding the current pixel by binary code.

Description

Halftone image representation

1 is a block diagram of a facsimile system for carrying out the present invention.

2 is a flow chart according to the present invention.

3 is a diagram illustrating a wave dither matrix according to the present invention.

4 is an exemplary view of a search window for boundary determination of the present invention.

5 is an explanatory diagram of data conversion according to the halftone image representation of the present invention.

6 is a memory map diagram for halftone image representation of the present invention.

7 is a diagram illustrating wave dithering according to the present invention.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a halftone image representation method of an image processing system, and in particular, to enhance the readability of a character by emphasizing the edge of the character and at the same time by dithering using a wave dither matrix. The present invention relates to a halftone image representation method enabling high-speed data transfer.

In general, image processing systems such as facsimile image scanners, printers, and copiers use a dither matrix to obtain halftone image representation from grayscale image data of manufactured originals such as photographs and drawings.

The halftone includes a real halftone and a dummy halftone, and their expression methods are also different. Since the present invention relates to the counterfeit half of the dummy, the description of the former case will be omitted and only the latter expression methods will be described. The method of expressing the halftone of the counterfeit includes a systematic dither method, a random dither method, an error distribution method, an average error minimum (error variance) method, an average value, a limit method, and a concentration pattern method.

Among the above expression methods, the systematic dither method determines the threshold value to be binarized from the position information of each pixel, and is the simplest method that does not change the threshold position in the matching matrix for the whole image. There are a dot intensive type such as scroll and the like and a dot dispersion type having high spatial frequency characteristics such as Bayer and Dot.

A conventional method of obtaining halftone image representation from a gradation image quantized with a constant gradation using the dither matrix as described above is disclosed in detail in 1990 Patent Application No. 16884 of the applicant and the same applicant.

However, the conventional halftone image representation method using the dither matrix does not consider human visual characteristics that are sensitive to the boundary between text and background, resulting in deterioration of image quality at the boundary portion and transfer time in transferring the obtained halftone image data. There was a problem with this lengthening.

Accordingly, an object of the present invention is to provide a halftone image representation method of grayscale image data in an image processing system, which can improve readability of a character by emphasizing the boundary between a character and a background in consideration of human visual characteristics. In providing.

Another object of the present invention is a method for expressing a halftone image of grayscale image data, the average run length of halftone image data using a wave dither matrix to shorten the transmission time of the halftone image data. It is to provide a halftone image representation method that can be increased).

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of a facsimile system for carrying out the present invention, in which a central processing unit (hereinafter referred to as CPU) 10 for controlling a system and performing various operations, and a document under the control of the CPU 10; An original scanning unit 20 for scanning and generating an analog image signal, converting the generated analog image signal into digital data, and outputting the image data output from the original scanning unit 20, and received image data and dither The CPU 10 stores the matrix in an internal memory and encodes and decodes the image data to and from another system, and the image data stored in the signal processor 30. The recording unit 40 which is inputted by the control unit to record on the recording sheet and the image data stored in the signal processing unit 30 is transferred to another system through the public circuit network under the control of the CPU 10. And it consists of a transmitting unit (50) for receiving from the other system.

The configuration of FIG. 1 is a block diagram of a general facsimile system. 2 is a flowchart according to the present invention, which includes a scanning process of converting an original into pixel data and converting the image into image data of a preset gray scale to store a current pixel value in a buffer, and searching for a current pixel (main pixel) of the scanning process. A local brightness difference determination process for determining a local brightness difference in a window; a boundary discrimination process for checking whether the local brightness difference is greater than a local threshold; and a current pixel is a boundary portion when the local brightness value difference is greater than a local threshold in the boundary discrimination process. If it is larger than the local brightness value, the image is half-weighted to "white", and if it is small, it is "black" to emphasize the boundary; if the current pixel value is larger than the local threshold, the half-tone image A wave dithering process, which binarizes to "white" and a small to "black", the boundary enhancement process and the wave dithering process After the completion of the scan is carried out to check whether the completion of the scan proceeds to the scanning process, and when it is completed, the end process is completed.

3 is a diagram illustrating a wave dither matrix according to the present invention, in which the corresponding threshold of 16 grays is continuity between pixels in a line in a unit matrix of 4 rows x 4 columns when the input image is 16 grays. It shows arrangement in wave form of 8 rows x 8 columns.

The wave dither matrix of FIG. 3 is stored in the second memory area of the signal processor 30, and the gray level of the corresponding threshold value, the row and the column may be changed according to the gray level of the input image.

4 is a diagram illustrating a search window for boundary determination according to the present invention. When the brightness values of pixels in the X-th line (row) and j-th column of the document are the current pixel value Pk, the current pixel value Pk is compared with the current pixel value Pk. The previous line pixel value PA-PC and the previous pixel value Pk 'of the current line are designated as a search window.

FIG. 5 is an explanatory diagram of data conversion according to the halftone image representation of the present invention. FIG. 5A is a scan of an original when the pixel P (i, j) of the ninth line and the jth column of the original is the current pixel value Pk. And previous pixel values P (i, 0), P (i, 1),... Of the current line in accordance with the binarization processing. P (i-1, j-1), P (i-1, j), P (i-1, j + of the pixel values P (i, j-2) and one line previous line i-1 1), P (i-1, j + 2),... , P (i-1, n-1) and P (i-1, n) are stored in the first memory area, which is a one-line memory of the signal processing unit 30. FIG. 5B shows data conversion when the next pixel P (i, j + 1) is set to the current pixel value Pk after completion of binarization processing for the pixel P (i, j) of FIG. 5A.

6 is a memory map for halftone representation of the present invention, where Tdif is a local threshold set in advance to determine whether a boundary is present with respect to the current pixel value Pk, and Pk is a brightness value of the current pixel, and Pk ' Is the previous pixel value in the current line with respect to the current pixel value Pk, and Bdif is the local brightness difference that is the difference between the maximum and minimum brightness values among the pixels in the search window for the current pixel and compared with the local threshold Tdif. A value for determining whether the current pixel is a boundary part, and Bv is a local brightness value obtained by arithmetically averaging the maximum brightness value and the minimum brightness value to binarize the current pixel value Pk when the current pixel is the boundary part.

The values of FIG. 6 are loaded and stored in the internal buffer of the CPU 10 from the signal processor 30.

FIG. 7 is an explanatory diagram of wave dithering using the wave dither matrix according to the present invention. FIG. 7a is an exemplary diagram of a grayscale image digitally converted into 16 grayscales by scanning an original, and FIG. 7b is an exemplary diagram of a conventional Bayer dither matrix. . FIG. 7C is an illustration of halftone representation when dithering using the Bayer dither matrix of FIG. 7B with respect to the grayscale image of FIG. 7A, and FIG. 7D is the wave of FIG. 3 of the present invention with respect to the grayscale image of FIG. It is an illustration of halftone representation when dithering using a dither matrix.

7E is an illustration of a line-shaped dinner matrix.

Hereinafter, an operation of an embodiment of the present invention will be described in detail with reference to FIGS. 1 to 7.

First, in Fig. 1, under the control of the CPU 10, the document scanning unit 20 scans an original, quantizes it into sampling and n bits, and converts the image into image data having a set gradation. At this time, if the input document is A4 size paper, the number of pixels per line becomes 1728 when the document scanning unit 20 samples at 200 dpi (dot per inch). The output of the document scanning unit 20 is input to the signal processing unit 30 and stored in a line memory having a storage area capable of storing one line of grayscale image data. The following description will be given on the assumption that the gray-scale image data is 4-bit quantized to have a gray scale of 0-15, and that the number of pixels in one line is 1728 as described above.

In addition, the CPU 10 binarizes and encodes the stored grayscale image data of the signal processing unit 30 and transmits the same to the public line network through the transmission unit 50.

On the other hand, the signal received by the transmission unit 50 from the public line network is decoded and reproduced by the signal processing unit 30 and reproduced on the recording paper by the recording unit 40.

If the system of FIG. 1 now starts the operation of scanning a gradated document such as a picture or a picture to obtain a halftone image representation, the CPU 10 initializes an internal buffer such as that of FIG. 6 and then (2a) in FIG. Perform the steps. At this time, the local threshold Tdif is loaded from the signal processor 30 into the internal buffer to clear the previous pixel value Pk ', the current pixel value Pk, the local brightness difference Bdif, and the local brightness value Bv. In step (2a), the document scanning unit 20 is controlled, and the document is copied in image units, and the current pixel (primary pixel) value Pk is stored in the same buffer as the sixth degree.

Next, in step (2b), a search window Sw equal to the fourth degree is set for the current pixel Pk, and the adjacent pixel value PA-PC in the search window Sw is set as the first signal of the signal processor 30. After loading from the memory area (line memory) into the buffer of FIG. 6, the maximum brightness value and the minimum brightness value of the pixels in the search window Sw are searched and the brightness difference between the searched maximum brightness value and the minimum brightness value is determined by calculation. And store it in the buffer as shown in FIG. 6b as the local brightness difference Bdif. In this case, the adjacent pixel value PA-PC is a brightness value of the adjacent pixel in the previous line that has been halftone-expressed, and if the current pixel is the first pixel of the first line of the original, the line memory and the CPU 10 of the signal processor 30 may be used. It becomes the value cleared according to the initialization of the internal buffer.

Next, in step 2c, it is determined whether the local brightness difference Bdif is greater than the local threshold value Tdif of FIG. 6, and if it is large, it is determined that the current pixel is in the boundary between the character and the background, and the process proceeds to step 2d. If it is small, determine that it is not the boundary part, and proceed to step (2h) to perform wave dithering.

In step (2d), the sum of the maximum brightness value and the minimum brightness value is divided by two to determine a local brightness difference (Bv) and stored in the same buffer as the sixth degree, and in step (2e ~ 2g), the current pixel brightness difference ( Checking whether Pk) is greater than the local brightness difference Bv, if it is large, binarizes the halftone image for the current pixel to "white", and if it is small, to "black" to emphasize the boundary of the character.

In step (2h), wave dithering is performed by checking whether the current pixel value Pk is larger than the corresponding threshold of the corresponding row and column of the wave dither matrix, such as the third degree, stored in the second memory area of the signal processor 30. do. At this time, if the brightness value Pk of the current pixel is larger than the corresponding threshold, the process proceeds to step (2g) to binarize the halftone image of the current pixel to "white", and to step (2f) to binarize to "black". . Here, as the corresponding threshold is wave dithering using the wave matrix having the continuity between the pixels of the line as shown in the third diagram, the run length of the half-tone image data which is binarized compared to the dithering using the dither matrix having the irregular gradation array is conventional. Longer

In other words, for example, halftones are represented in the grayscale image of FIG. 7a by using the conventional Bayer dither matrix, such as the 7b diagram. It can be seen that the halftone is expressed as the 7d degree when dithering using x4.

Therefore, as the average run length of black and white is increased, the transfer time of half-tone image representation data is greatly shortened. According to the experimental results, the transfer time can be shortened to about 1/3. In addition, since the wave dither matrix is arranged in a wave form in each of the four rows x four columns of the eight rows x eight columns, the image according to the halftone expression becomes a wave shape, which is more effective in the human eye than a conventional Bayer dither matrix. It is possible to express more natural parts with uncomfortable intermediate brightness values, which can greatly improve the visual quality.

And when dithering by a simple lyric dither matrix such as the 7e diagram, it is difficult to read the character on a slightly dark background, whereas the dithering by the wave dither matrix does not have continuity with adjacent lines of each line. Since the halftone representation of the is distinguished from the horizontal stroke of the character, the readability of the character can be greatly improved.

After the halftone expression is obtained for the current pixel by performing the steps (2f) and (2g), the signal processor may convert the previous pixel value Pk 'to process the next adjacent line which has not yet been scanned in steps 2i-2j. The current pixel value Pk is stored as the previous pixel value Pk 'in the buffer of FIG. 6 for processing of the next pixel.

Thereafter, if the document has been scanned for completion in step (2k) and is not completed, the next step is processed by looping to step (2a) and repeating the above-described process, and the process ends when document scanning is completed.

In the above-described description of the present invention, a facsimile system has been described as an example, but it can be applied to an image processing system that performs halftone image representation such as an image scanner, a printer, a copier, and the like.

As described above, the present invention, in the halftone representation method of the grayscale image data of the image system, not only improves the image quality of the halftone image by emphasizing a boundary sensitive to human visual characteristics, but also especially in a document having mixed characters. By emphasizing the boundary of, the readability of the characters can be greatly improved.

In addition, by using the wave dither matrix, the average run length is much longer when dithering than when using the conventional dither matrix. Therefore, the transmission time can be shortened, and the input image having 16 gradations can be expressed as almost 64 gradations. .

Claims (5)

A halftone representation method of gradation image data of an image processing system, comprising: a scanning process for converting an original into pixel data and converting the image into image data of a preset gradation and storing a current pixel value in a buffer; A local brightness difference determination process for determining a local brightness difference in a search window; a boundary discrimination process for checking whether the local brightness difference is greater than a local threshold; and a current pixel being a boundary portion when the local brightness difference is greater than a local threshold in the boundary discrimination process. If it is larger than the local brightness value, the halftone image is binarized to "white" and if small, it is binarized to "black" to emphasize the boundary; and if the local brightness difference is less than the local threshold in the boundary discrimination process, the current pixel If it is determined that it is not a boundary and is larger than the corresponding threshold of the wave dither matrix, the halftone image is set to "white". A halftone image representation method comprising a wave dithering process that is binarized and small if it is "black". The method of claim 1, wherein the halftone expression method further checks whether or not the document has been scanned after the boundary enhancement process and the wave dithering process. Halftone image representation method comprising a. The halftone image representation method of claim 1, wherein the wave dither matrix of the wave dithering process has continuity between corresponding thresholds belonging to each line, and the corresponding thresholds of each line are arranged in a wave form. 4. The method of claim 3, wherein the wave dither matrix has no corresponding thresholds of each line continuity with corresponding thresholds of adjacent lines. 5. The halftone image representation method according to claim 4, wherein said wave dither matrix is composed of eight rows x eight columns corresponding to sixteen grayscale levels corresponding to sixteen grayscale input grayscale image data.