KR20000065587A - Image binarization method for removing the noise due to isolated pixel - Google Patents

Abstract

PURPOSE: A local binarization method removing an isolation point noise is to remove an isolation point noise of key pixels using a binarization result of adjacent pixels adjacent to the key pixels, thereby improving the picture quality of a binarization image. CONSTITUTION: The local binarization method removing an isolation point noise comprises the steps of: creating a binarization result image by performing a binarization of respective input images in a binarization object image; orderly inputting notable binary pixels from the binarization result image; determining whether all of adjacent binary pixel groups placed in eight directions centered at the notable binary pixels become binarized in the same binary value; when it is determined that the adjacent binary pixel group become binarized in the same binary value, deciding the notable binary pixel as an isolation point pixel and determining whether the same binary value is white or black; and when it is determined that the same binary value is white, substituting the white pixel for the isolation pixel and when it is determined that the same binary value is black, substituting the black pixel for the isolation pixel.

Description

Image binarization method for removing the noise due to isolated pixel}

The present invention relates to a localized binarization method in which isolated point noise has been removed. More particularly, after detecting isolated point noise according to the binarization result of adjacent pixels adjacent to the pixel of interest, the isolated point noise is detected using the binarization result of adjacent pixels. The image binarization method removes the isolated point noise which improves the image quality of a binarized image by eliminating a.

One of the typical imaging systems, the scanner (ie, reading means) is the most common means for reading printed texts, photographs, human-written letters or pictures, and includes a multifunction device, a document translator, and CAD (Computer Aided Design). Essential components such as computer, facsimile, character recognizer and digital copier.

In recent years, as office electronics, which are at the crossroads of development, is rapidly increasing in demand for office automation devices such as printers, scanners, and facsimile machines, each office automation device is developed with high performance to expand its own functions. In addition, it is a trend to produce and provide a product that performs a complex document output function at the same time to reduce the economic burden and installation space for the user by developing each office automation equipment that was used independently as an integrated.

In particular, in devices having a scanner capable of reading a document therein, such as a facsimile machine or a multifunction printer, after reading an image having multiple levels of gradation through the scanner, the characteristics of the binarization method of the read data are characterized by In evaluating performance, it becomes a major evaluation item.

In general, "image binarization" refers to a gray level image represented by a multi-step gray level in an image system, where a quantization number is one bit, that is, a binary number of '1' and '0'. An image processing technique used to express a binary image or binary image. Image compression is performed based on a signal processing characteristic capable of preserving the characteristic information of a target image while significantly reducing the amount of processing data. It is widely used in the field of image output, image recognition, etc.

Due to this wide range of applications, various image binarization methods are known.

The most common binarization method is to obtain a binarized image by determining an arbitrary threshold, assigning " 1 " to pixel values larger than this threshold and " 0 " to pixel values smaller than the threshold. Thus, image binarization processing is a kind of threshold processing for classifying gray levels up and down based on a threshold value. The most important thing in the threshold processing is the determination of a threshold value. There are also ways to determine more precise thresholds through concentration histograms.

On the other hand, in order to properly binarize the edge regions of the visually important image, there is a method of performing binarization after emphasizing the outline of the original image through a predetermined edge enhancement filter. It is known that a good binarization image can be obtained as compared with the method that does not include the emphasis process.

An example thereof will be described with reference to FIG. 1.

For example, when the gradation value of the original image is given as shown in Table 1,

78 77 76 93 152 176 227 222

When spatial filtering is performed using a 1 × 3 edge enhancement mask as shown in Table 2,

-One 3 -One

You can get the result with the outline outlined in Table 3.

77 58 51 187 149 255

Fig. 1 shows the gradation value of the original image and the gradation value of which the outline is highlighted, where the solid line represents the gradation value of the original image, and the dotted line represents the gradation value of the image as a result of emphasizing the outline.

At this time, if the threshold value is set to 150, 78, 58, 51 in Table 3 are set to "0", 187 is set to "1", 149 is set to "0", and 255 is set to "1". When the pixel set to "" is shown in white and the pixel set to "0" is shown in black, two lines of contour lines are formed as there are one contour lines on both sides of 149, 149, and 255 where the contour lines are distributed. Appearance results in deformation or distortion of the original image.

As described above, even when the binarization method including the contour enhancement process, which is known to have superior binarization characteristics, is used, the edge region and the edge region are generated as a false boundary occurs in the contour region as described with reference to FIG. There is an expectation that a localized binarization method capable of performing binarization by separating non-edge regions is expected. Accordingly, many localized binarization methods are known.

Applicants of the present invention have also proposed a number of localized binarization methods, including Korean Patent Application No. 91-8508, "Digitization Method Using Local Maximum and Minimum Brightness Values" in order to meet such technical expectations.

Republic of Korea Patent Application No. 91-8508, "binarization method using local maximum and minimum brightness values", as the name of the invention means using a binarization mask having a predetermined window centered around the pixel of interest in the image After masking, the pixels of interest are adaptively binarized using the local maximum and minimum values of the pixels masked in this binarization mask.

The conventional minimum-to-minimum localized binarization method includes a scanning step of scanning a scanning target image in a multi-level gray scale based on a scanning method, and using a binarization mask having a predetermined window according to a scanning progress direction. A binarization masking step of masking adjacent pixels centering on the pixel of interest; obtaining a maximum value and a minimum value of pixels masked in the binarization mask; and then, if the absolute value of the difference between the maximum value and the minimum value is smaller than a predetermined reference value, Binarize the pixel of interest using a predetermined global threshold and otherwise binarize the pixel of interest using a predetermined local threshold that is relatively smaller than the global threshold.

However, the binarization methods according to the prior art, including the maximum-minimum based local binarization method, which is one of the relatively advanced binarization methods in the prior art, devote a separate routine to reduce image quality degradation caused by the isolated point. Spot noise, which causes visual rejection, is emphasized in the extracted binarized image as it is not present, which is not only a major factor in deteriorating the image quality of binarized images, but also compression at run length coding. There is a problem such as causing a decrease in efficiency.

Accordingly, the present invention has been devised in view of such a problem, and since the isolated point noise is detected according to the binarization result of adjacent pixels adjacent to the pixel of interest, the binarized image is removed by removing the isolated point noise using the binarization result of adjacent pixels. It is an object of the present invention to provide an image binarization method that removes the isolated point noise to improve the image quality.

1 is a graph showing a gray scale value of an original image and a gray scale value of which an outline is emphasized according to the prior art;

2 is a block diagram showing a preferred embodiment of the image binarization method removing the isolated point noise according to the present invention;

3a is a typical three Figure showing an example of three binarization masks,

3b is a typical five 5 shows an example of a binarization mask.

4 is an exemplary diagram for describing an adjacent binary pixel group of the present invention.

In order to achieve the above object, the image binarization method of removing the isolated point noise according to the present invention includes: an image binarization step of generating a binary result image by performing image binarization on each pixel input from a binarization target image; A pixel input step of sequentially receiving a binary pixel of interest in the binary result image; An isolated point determination step of determining whether adjacent binary pixel groups located in eight directions around the binary binary pixel are binarized with the same binary value; If the adjacent binary pixel groups are all binarized to the same binary value, determining the binary pixel of interest as the isolated point pixel, and then determining whether the same binary value is white or black; And an isolated point pixel replacing step of replacing the isolated point pixel with white when the same binary value is white, and otherwise replacing the isolated point pixel with black.

The image binarizing may include sequentially specifying a pixel of interest to be binarized in the binarization target image; Comparing and determining whether the pixel of interest is greater than or equal to a predetermined upper reference threshold; Binarizing the pixel of interest to white if the pixel of interest is greater than or equal to the upper reference threshold; Comparing and determining whether the pixel of interest is less than or equal to a predetermined lower reference threshold when the pixel of interest is smaller than the upper reference threshold; Binarizing the pixel of interest to black if the pixel of interest is less than or equal to the lower reference threshold; When the pixel of interest is smaller than the upper reference threshold and larger than the lower reference threshold, the adjacent pixels are masked around the pixel of interest using a binarization mask having a preset window size, and then the maximum gray level value in the masked pixel group is determined. Comparing and calculating a minimum gray value; Calculating a maximum gradation that is even a gradation between the maximum gradation value and the minimum gradation value to determine whether even the maximum gradation is greater than a predetermined threshold; Binarizing the pixel of interest by averaging the maximum gray value and the minimum gray value if the maximum gray level is even greater than or equal to the threshold value; And if the maximum system is even smaller than the threshold, binarizing the pixel of interest to white.

Hereinafter, with reference to the accompanying drawings a preferred embodiment of the image binarization method of removing the isolated point noise according to the present invention will be described.

Fig. 2 is a block diagram showing a preferred embodiment of the image binarization method by removing the isolated point noise according to the present invention.

According to a preferred embodiment of the image binarization method which removes the isolated point noise according to the present invention, as shown in FIG. 2, an image binarization is performed by performing image binarization on each pixel input from a binarization target image to generate a binary result image. Step S100;

A pixel input step (S200) of sequentially receiving a binary pixel of interest in the binary result image;

An isolated point determination step (S300) of determining whether all adjacent binary pixel groups located in eight directions about the binary pixel of interest are binarized with the same binary value;

As a result of the determination of the isolated point determining step (S300), when all of the adjacent binary pixel groups are binarized to the same binary value, after determining the binary pixel of interest as the isolated point pixel, determining whether the same binary value is white or black. Step S400; And

If the same binary value is white, the isolated pixel is replaced with white; otherwise, the isolated pixel substitution step S500 of replacing the isolated pixel with black is performed.

The image binarizing step (S100) may include: sequentially specifying a pixel of interest to be binarized in the binarization target image (S110);

Comparing and determining whether the pixel of interest is greater than or equal to a predetermined upper reference threshold (S120);

If the pixel of interest is greater than or equal to the upper reference threshold, binarizing the pixel of interest to white (S130);

If the pixel of interest is smaller than the upper reference threshold, comparing and determining whether the pixel of interest is less than or equal to a predetermined lower reference threshold (S140);

If the pixel of interest is less than or equal to the lower reference threshold, binarizing the pixel of interest to black (S150);

When the pixel of interest is smaller than the upper reference threshold and larger than the lower reference threshold, the adjacent pixels are masked around the pixel of interest using a binarization mask having a preset window size, and then the maximum gray level value in the masked pixel group is determined. Comparing and calculating a minimum gray value (S160);

Calculating even a maximum gradation that is even a gradation between the maximum gradation value and the minimum gradation value to determine whether the maximum gradation is even greater than a predetermined threshold (S170);

If even the maximum gray is even greater than or equal to the threshold, binarizing the pixel of interest by averaging the maximum gray value and the minimum gray value (S180); And

If even the maximum system is even smaller than the threshold, it is configured to binarize the pixel of interest to white (S190).

Hereinafter, a process of performing a preferred embodiment of the image binarization method removing the isolated point noise according to the present invention configured as described above will be described in detail with reference to the accompanying drawings.

3a is a typical three 3 is a diagram illustrating a three binarization mask, and FIG. 3B is a typical five. 5 is an exemplary diagram illustrating a five binarization mask, and FIG. 4 is an exemplary diagram for describing an adjacent binary pixel group of the present invention.

First, in the image binarization step S100, image binarization is performed on each pixel input from the binarization target image to generate a binary result image.

Note that the present invention is not dependent on the image binarization technique for generating a binary result image. In other words, even if a binary result image is subjected to binarization by applying any image binarization technique, a noise removal process according to detection of an isolated pixel may be performed.

However, in order to help the understanding of the present invention, a preferred embodiment of the image binarization step S100 will be described below.

First, in step S110, pixels of interest to be binarized in the binarization target image are sequentially designated.

Subsequently, in step S120, the pixel of interest is compared and determined to be greater than or equal to a predetermined upper reference threshold.

As a result of the determination in step S120, if the pixel of interest is greater than or equal to the upper reference threshold, the pixel of interest is binarized to white in step S130. On the other hand, if the pixel of interest is smaller than the upper reference threshold, step S140 compares and determines whether the pixel of interest is less than or equal to a predetermined lower reference threshold.

As a result of the determination in step S140, if the pixel of interest is less than or equal to the lower reference threshold, the pixel of interest is binarized to black in step S150.

Here, the upper reference threshold and the lower reference threshold correspond to a kind of global threshold. A pixel of interest having a gray scale value of a predetermined size or more (that is, a high reference threshold) is binarized to white and a gray scale value of a predetermined size or less ( In other words, the pixel of interest having the lower reference threshold) is binarized to black. This may have the side effect of enhancing the noisy isolated point pixel as binarization is performed without considering locality of the pixel of interest. The present invention is intended to remove such unnecessarily noisy isolated point noise.

On the other hand, if the pixel of interest is smaller than the upper reference threshold and larger than the lower reference threshold as a result of the determination of steps S120 and S140, in step S160 as shown in FIG. 3 or 5 After masking adjacent pixels around the pixel of interest using a binarization mask having a predetermined window size of 5, the maximum and minimum gray values are compared and calculated in the masked pixel group.

In step S170, even the maximum system that is even the system between the maximum gray value and the minimum gray value is calculated to determine whether even the maximum system is greater than the preset threshold.

As a result of the determination in step S170, if even the maximum system is greater than or equal to the threshold, in step S180, the pixel of interest is binarized by averaging the maximum gray value and the minimum gray value. That is, if the pixel of interest is larger than or equal to the average value of the maximum gray value and the minimum gray value, it is binarized to white, otherwise it is binarized to black.

On the other hand, if it is determined in step S170 that even the maximum system is smaller than the threshold, in step S190, the pixel of interest is binarized to white to remove a local background.

After acquiring the binary result image through the above process, in step S200, the binary pixel of interest is sequentially input from the binary result image.

Subsequently, in the isolated point determination step (S300), it is determined whether all adjacent binary pixel groups located in eight directions around the binary pixel of interest are binarized with the same binary value.

Referring to FIG. 4, it is determined whether adjacent binary pixel groups P1, P2, P3, P4, P5, P6, P7, and P8 are all binarized to the same binary value (ie, black or white) with respect to the binary pixel C of interest. To judge. Often, because of the nature of the image, such an isolated pixel is likely to be noise, so it is intended to detect and remove it.

As a result of the determination of the isolated point determining step (S300), if all of the adjacent binary pixel groups are binarized to the same binary value, in step S400, after determining the binary pixel of interest as the isolated point pixel, whether the same binary value is white or black. Judge.

As a result of the determination in step S400, if the same binary value is white, in the isolated pixel replacement step (S500), the isolated pixel is replaced with white (S510), otherwise, the isolated pixel is replaced with black (S520). It is possible to obtain a binary result image by removing the noisy isolated point pixels.

The terminologies used herein are terms defined in consideration of functions in the present invention, and may vary according to the intention or custom of a person skilled in the art, and the definitions are based on the contents throughout the present application. Will have to be lowered.

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. As modifications may be made, it is obvious that both the embodiments and modifications cited in the foregoing description belong to the claims of the present invention.

As described in detail above, an image binarization step of generating a binary result image by performing image binarization on each pixel input from the binarization target image; A pixel input step of sequentially receiving a binary pixel of interest in the binary result image; An isolated point determination step of determining whether adjacent binary pixel groups located in eight directions around the binary binary pixel are binarized with the same binary value; If the adjacent binary pixel groups are all binarized to the same binary value, determining the binary pixel of interest as the isolated point pixel, and then determining whether the same binary value is white or black; And an isolated point pixel replacing step of replacing the isolated point pixel with a white color if the same binary value is white, and otherwise replacing the isolated point pixel with a black color. According to this, there is an advantage that the quality of the binarized image can be improved by detecting the isolation point noise according to the binarization result of the adjacent pixel adjacent to the pixel of interest and then removing the isolation point noise using the binarization result of the adjacent pixel.

Claims (3)

An image binarization step of performing image binarization on each pixel input from the binarization target image to generate a binary result image; A pixel input step of sequentially receiving a binary pixel of interest in the binary result image; An isolated point determination step of determining whether all adjacent binary pixel groups located in a plurality of predetermined directions around the binary binary pixel are binarized with the same binary value; If the adjacent binary pixel groups are all binarized to the same binary value, determining the binary pixel of interest as the isolated point pixel, and then determining whether the same binary value is white or black; And And removing the isolated point noise if the same binary value is white, replacing the isolated point pixel with white, otherwise, replacing the isolated point pixel with black. The method of claim 1, wherein the predetermined plurality of directions are: Localized binarization method which removes isolation point noise characterized by eight directions. The method of claim 1, wherein the image binarization step, Sequentially specifying a pixel of interest to be binarized in the binarization target image; Comparing and determining whether the pixel of interest is greater than or equal to a predetermined upper reference threshold; Binarizing the pixel of interest to white if the pixel of interest is greater than or equal to the upper reference threshold; Comparing and determining whether the pixel of interest is less than or equal to a predetermined lower reference threshold when the pixel of interest is smaller than the upper reference threshold; Binarizing the pixel of interest to black if the pixel of interest is less than or equal to the lower reference threshold; When the pixel of interest is smaller than the upper reference threshold and larger than the lower reference threshold, the adjacent pixels are masked around the pixel of interest using a binarization mask having a preset window size, and then the maximum gray level value in the masked pixel group is determined. Comparing and calculating a minimum gray value; Calculating a maximum gradation that is even a gradation between the maximum gradation value and the minimum gradation value to determine whether even the maximum gradation is greater than a predetermined threshold; Binarizing the pixel of interest by averaging the maximum gray value and the minimum gray value if the maximum gray level is even greater than or equal to the threshold value; And And if said maximum system is even less than said threshold, binarizing said pixel of interest to white to produce said binary result image.