KR100242298B1 - Image processing method for converting gray-scale image of text document into binary image - Google Patents

Abstract

end. The technical field to which the invention described in the claims belongs

A method of binarizing image data of multi-level gradations obtained by scanning a text document.

I. The technical problem to be solved by the invention

Provides a binarization method of a text document image that can improve image quality of a text boundary.

All. Summary of Solution of the Invention

If the pixel of interest is the boundary portion of the character, the mean value of the pixels of interest is compared with the average value of adjacent pixels in the edge direction around the pixel of interest, and the pixel of interest is binarized according to the comparison result.

la. Important uses of the invention

Used to binarize text document images.

Description

Image data binarization method of text document

The present invention relates to a method for binarizing image data obtained by scanning a document, and in particular, a method for binarizing image data having multi-value gray-scale obtained by scanning a text document into binary image data. It is about.

BACKGROUND ART Image input devices employed in facsimile, scanner, digital copier, and the like are typically processed after converting an analog image signal obtained by scanning a document into digital image data. At this time, the analog image signal is converted into digital image data having a gradation corresponding to a document image for each pixel. Since the gradation has a gradation, the amount of data is very large. For example, in the case of 256 gray levels, each pixel is represented by 8 bits. Accordingly, by binarizing the image data according to the scan, the amount of data is drastically reduced and then stored, transmitted or printed out. The binarization determines each pixel having a gradation as a ″ white ″ pixel or a ″ black ″ pixel, where ″ white ″ pixels and ″ black ″ pixels are relatively 1 of ″ 1 ″ or ″ 0 ″. It is represented by bit data.

This binarization adopts a different method depending on the type of document image. The document image is classified into a text document consisting of text, a picture document consisting of a picture (or a picture), and a document in which text and pictures are mixed.

As an example of a method that has been used to binarize a text document before, a description of a local allocation threshold algorithm (LAT) is as follows. First, a pixel value of a target pixel, which is a pixel to be binarized in grayscale image data, is compared with a maximum threshold Tmax. The maximum threshold Tmax is a preset threshold for binarizing the pixel of interest to ″ back ″ when the pixel value of interest is greater than the maximum threshold Tmax. At this time, if the pixel value of interest is larger than the maximum threshold value Tmax, the pixel of interest is binarized to ″ white ″. In contrast, if the pixel value of interest is not greater than the maximum threshold Tmax, the local brightness difference Wmax-Wmin is compared with the brightness difference threshold Tdiff. The local brightness difference Wmax-Wmin designates a window of constant size including the surrounding pixels for the pixel of interest, obtains the difference between the maximum value Wmax and the minimum value Wmin by obtaining the maximum value Wmax and the minimum value Wmin among the pixels in the window. will be. For example, the window is a 3x3 window, i.e., a window for designating pixels of interest and pixels in up, down, left, right, and diagonal directions with respect to the pixel of interest. The brightness difference threshold Tdiff is a threshold set in advance to identify an edge portion that is a boundary portion of the character, and the edge portion has a local brightness difference Wmax-Wmin larger than the brightness difference threshold Tdiff.

If the local brightness difference Wmax-Wmin is greater than the brightness difference threshold Tdiff, that is, the pixel of interest is the boundary portion of the character, the pixel value of interest is compared with the local brightness value (Wmax + Wmin) / 2. The local brightness value (Wmax + Wmin) / 2 corresponds to the average brightness value in the window. At this time, if the pixel value of interest is larger than the local brightness value (Wmax + Wmin) / 2, the pixel of interest is binarized to ″ white ″, otherwise it is binarized to ″ black ″. If the local brightness difference Wmax-Wmin is not greater than the brightness difference threshold Tdiff, the pixel value of interest is compared with the minimum threshold Tmin. The minimum threshold Tmin is a preset threshold for binarizing the pixel of interest to ″ black ″ when the pixel value of interest is less than the minimum threshold Tmin. If the pixel value of interest is larger than the minimum threshold Tmin, the pixel of interest is binned to ″ white ″, otherwise the pixel of interest is binarized to ″ black ″.

When binarizing image data with gray scale by the above-described method, edge direction is not taken into consideration as the pixel value of interest in the edge portion is simply compared with the local brightness value (Wmax + Wmin) / 2 so that the direction and continuity of the edge is not taken into account. This will be lumped. As a result, the image quality of the binarized image was deteriorated.

As described above, the conventional method for binarizing a character document image has a disadvantage in that the image quality of the digitized image is deteriorated due to agglomeration of the boundary portions of the character as it is binarized without considering the directionality and continuity of the edge.

Accordingly, an object of the present invention is to provide a binarization method of a character document image capable of improving image quality of a boundary portion of a character.

1 is a block diagram of a conventional scanner to which the present invention is applied;

2 is a process flow diagram according to an embodiment of the present invention.

According to the present invention for achieving the above object, when the pixel of interest is the boundary portion of the character, the average value of the adjacent pixels in the edge direction of the pixel of interest is compared with the pixel of interest, and the pixel of interest is binarized according to the comparison result. It features.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Many specific details are set forth in the following description and in the accompanying drawings to provide a more general understanding of the invention. It will be apparent to those skilled in the art that the present invention may be practiced without these specific details. And a detailed description of known functions and configurations that may unnecessarily obscure the subject matter of the present invention will be omitted.

1 shows a block diagram of a conventional scanner as an example of an apparatus to which the present invention is applied. The central processing unit 100 detects the document mounting by the document detection sensor 112 and controls the image processor 104 to scan the document when a scan start command is input from the operation panel 110. The scanning unit 102 includes an optical system for scanning an original and scans the original to generate an analog image signal. An image processor 104 drives the scan unit 102 to convert the image signal output from the scan unit 102 into digital data, and to process and binarize the image. The ROM (Read Only Memory) 106 stores the operation program of the CPU 100 and various reference data in advance. The random access memory (RAM) 108 temporarily stores data according to an operation of the CPU 100. The operation panel 110 provides a key input by the user to the CPU 100. The document detection sensor 112 detects that the document is mounted and notifies the central processing unit 100.

2 is a flowchart illustrating a process of binarizing one pixel for convenience, according to an exemplary embodiment of the present invention. The same applies to all pixels. The operation according to the flowchart of FIG. 2 is programmed to be performed by the image processor 104 of FIG.

In FIG. 1, as the document is detected by the document sensor 112 and a scan start command is input from the operation panel 110, the image processor 104 is controlled by the central processing unit 100. When the original is scanned by 102, an image signal corresponding to the original scan is output from the scanning unit 102. Then, the image processor 104 converts the image signal output from the scanning unit 102 into digital image data having a gray level, and stores the pixel value in the RAM 108 in units of lines to be scanned, and according to the flowchart of FIG. 2. Perform binarization.

First, in step 200 of FIG. 2, the image processor 104 compares the pixel value of interest of the image data with the maximum threshold Tmax as described above. If the pixel value of interest is greater than the maximum threshold value Tmax, the image processor 104 binarizes the pixel of interest to " white " In contrast, if the pixel value of interest is not greater than the maximum threshold Tmax, the image processor 104 compares the local brightness difference Wmax-Wmin with the brightness difference threshold Tdiff in step 204 to determine whether the pixel of interest is an edge, that is, a boundary of a character. Check it. The local brightness difference Wmax-Wmin and the brightness difference threshold Tdiff are as described above.

If the local brightness difference Wmax-Wmin is greater than the brightness difference threshold Tdiff, i.e., if the pixel of interest is the boundary portion of the character, the image processor 104 performs two adjacent pixels in the edge direction around the pixel of interest in step 206. Compare the average value of (I1 + I2) / 2 with the pixel value of interest. Since detecting the edge direction is a conventional method, detailed description thereof will be omitted. I1 and I2 are two pixel values adjacent to the pixel of interest. If the pixel value of interest in step 206 is not smaller than the average value (I1 + I2) / 2, the image processor 104 binarizes the pixel of interest to ″ white ″ in step 208. In contrast, if the pixel value of interest is smaller than the average value (I1 + I2) / 2, the image processor 104 binarizes the pixel of interest to ″ black ″ in step 210.

Therefore, when the pixel of interest is the boundary portion of the character, the pixel is binarized by applying the average value of adjacent pixels in the edge direction to binarize it according to the directionality and continuity of the edge. This prevents agglomeration of the boundary of characters.

On the other hand, if the local brightness difference Wmax-Wmin in step 204 is not greater than the brightness difference threshold Tdiff, i.e., it is not the boundary portion of the character, the image processor 104 performs the same operation as in the normal operation in steps 201 to 216. It makes sense. That is, the image processor 104 compares the pixel value of interest with the minimum threshold value Tmin in step 212. If the pixel value of interest is greater than the minimum threshold value Tmin, the image processor 104 binarizes the pixel of interest to " back " On the other hand, if the pixel value of interest is not larger than the minimum threshold Tmin, the image processor 104 binarizes the pixel of interest to " black "

By the operation as described above, binarization is performed on one pixel, and when it is finished, the same operation is repeatedly performed on the next pixels until it is returned to complete the binarization of the entire image data. .

Meanwhile, in the above description of the present invention, specific embodiments have been described, but various modifications can be made without departing from the scope of the present invention. In particular, the embodiment of the present invention shows an example of applying the present invention to a scanner, but the same applies to binarizing image data of a text document. In the case of the boundary portion of the character, an example in which the pixel value of interest is compared with an average value of two adjacent pixels in the edge direction is illustrated. However, the number of adjacent pixels that take an average value may vary as necessary. Therefore, the scope of the invention should not be defined by the described embodiments, but should be defined by the equivalent of claims and claims.

As described above, when the pixel of interest is the boundary portion of the character, the image quality is improved by preventing binarization of the boundary portion of the character by binarizing according to the direction and continuity of the edge.

Claims (3)

In the method for binarizing image data having multilevel gradations obtained by scanning a text document, If the pixel of interest of the image data is a boundary of a character, comparing the pixel value of the pixel with the mean value of adjacent pixels in the edge direction with respect to the pixel of interest; And binarizing the pixel of interest according to a result of comparison between the pixel of interest value and the average value. According to claim 1, wherein the binarization process, If the pixel value of interest is smaller than the average value, binarizing the pixel of interest to ″ black ″; And binarizing the pixel of interest to ″ white ″ if the pixel of interest value is not less than the average value. In the method for binarizing image data having multilevel gradations obtained by scanning a text document, Comparing the pixel value of interest of the image data with a preset maximum threshold; Binarizing the pixel of interest to ″ white ″ if the pixel value of interest is greater than the maximum threshold; Checking whether the pixel of interest is a boundary portion of a character if the pixel of interest value is not greater than the maximum threshold value; If the pixel of interest is a boundary portion of a character, comparing the pixel value of the pixel with the mean value of adjacent pixels in an edge direction with respect to the pixel of interest; If the pixel value of interest is smaller than the average value, binarizing the pixel of interest to ″ black ″; Binarizing the pixel of interest to ″ white ″ if the pixel of interest value is not less than the average value; Comparing the pixel value of interest with a preset minimum threshold value if the pixel of interest is not a boundary of a character; Binarizing the pixel of interest to ″ white ″ if the pixel value of interest is greater than the minimum threshold value; And binarizing the pixel of interest to " black " if the pixel of interest value is not greater than the minimum threshold.

Images