KR19980066129A - How to distinguish between text and picture areas in binarized mixed images - Google Patents

Abstract

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

The present invention relates to a method for distinguishing a text area and a picture area in a binarized mixed image.

I. The technical problem that the invention is trying to solve

Distinguish between the letter area and the picture area in the binarized mixed image.

All. Summary of Solution of the Invention

The adjacent pixel grouping step of performing adjacent pixel grouping using the value of the black pixel as the center pixel value in the region of the binarized mixed image, and searching the number of black pixel groups and the maximum black pixel group area after the adjacent pixel grouping. The black pixel group number and the maximum black pixel group area searching step are divided into picture regions when the black pixel group number is larger than one specific threshold and the maximum black pixel group area is smaller than another specific threshold. In this case, it is characterized in that it comprises an image classification step of dividing into a letter area.

la. Important uses of the invention

Used to process image data.

Description

How to distinguish between text and picture areas in binarized mixed images

The present invention relates to a method of processing image data, and more particularly, to a method of classifying image data.

Conventionally, when enlarging or reducing the binarized mixed image, the letter area and the picture area in the binarized image are not distinguished, but are simply enlarged or reduced according to a certain ratio. In more detail, when the central processing unit provides the image processing unit with an integer magnification and the integer magnification to enlarge or reduce, the image processing unit reduces or enlarges the binarized mixed image according to the integer magnification. As the image is reduced or enlarged as described above, the edge portion of the mixed image is severely distorted, and only the edge portion is smoothed.

In enlarging or reducing the binarized mixed image as described above, the font area of the binarized mixed image has less distortion than the picture area. However, in the picture area where the binarization technique is complicated, the black dots are entangled and the distortion is severe when the zooming in or out is performed in the above manner. Accordingly, when enlarging or reducing the binarized image, the enlargement or reduction of the binarized image was required. However, in the case of a mixed image, it is difficult to enlarge or reduce according to the image because the picture and the font are mixed.

As described above, in the enlargement or reduction of the conventionally mixed image, there is a problem in that distortion of the image of the picture region among the mixed image is severe.

Accordingly, an object of the present invention is to provide a method of distinguishing a text area from a picture area in a binarized mixed image which improves the distortion of an image of the picture area when the binarized mixed image is enlarged or reduced.

1 is a block diagram of a conventional facsimile;

2 is a flowchart illustrating a method of distinguishing a text area from a picture area in a binarized mixed image according to a preferred embodiment of the present invention.

The present invention for achieving the above object is an adjacent pixel grouping step of performing adjacent pixel grouping with the value of the black pixel as the center pixel value in a region of a predetermined size of the binarized mixed image, and after the adjacent pixel grouping, the black pixel A black pixel group number and a maximum black pixel group area search step for searching the number of groups and the maximum black pixel group area, and the black pixel group number is larger than one specific threshold and the maximum black pixel group area is smaller than another specific threshold. In this case, it is characterized by including an image classification step of dividing into a picture area, and otherwise dividing into a picture area.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description and the annexed drawings, numerous specific details are set forth in order to provide a more thorough understanding of the present 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.

The present invention can be applied to a device equipped with a scanner or to any device that enlarges or reduces the binarized mixed image. Hereinafter, an example in which the present invention is applied to a facsimile, which is one of apparatuses including a scanner and enlarging or reducing the binarized mixed image, will be described.

1 illustrates a block diagram of a conventional facsimile, and the controller 10 controls the overall operation of the facsimile according to a preset program. In particular, the controller 10 executes a program according to a method of distinguishing a text area from a picture area in a binarized mixed image according to a preferred embodiment of the present invention. The modem 12 modulates the image data output to the image processing unit 26 in the transmission mode in the transmission mode under the control of the control unit 10, and demodulates the image data in the reception mode. The network control unit (NCU) 14 forms a transmission / reception path between the telephone line and the modem 12 under the control of the control unit 10. The printer 16 prints an image on paper under the control of the control unit 10. The memory 18 includes a program memory for storing a program and a data memory for temporarily storing data generated during the execution of the program. The scanner 20 scans an original under the control of the control unit 10. The operation panel 22 includes a plurality of keys and a display device for interfacing between the user and the control unit 10. The sensor unit 24 senses various states of the facsimile, and provides the sensing data accordingly to the controller 10. The image processor 26 receives image data under the control of the controller 10 and performs encoding and decoding.

2 is a flowchart illustrating a method of distinguishing a text area from a picture area in a binarized mixed image according to a preferred embodiment of the present invention.

In step (28) of FIG. 2, the controller 10 searches whether to enlarge or reduce the binarized mixed image and distinguishes the picture area and the text area from the binarized mixed image when the binarized mixed image is enlarged or reduced. If not, it is determined that the picture area and the text area are not distinguished in the binarized mixed image. In this case, if it is determined that the picture area and the text area of the binarized mixed image are distinguished, the controller 10 performs step (30); otherwise, performs step (46). In step 46, the controller 10 performs a corresponding operation.

In step (30), the controller 10 searches whether the classification of the text area and the picture area of the entire binarized mixed image is finished. At this time, the controller 10 terminates the processing program according to the flowchart when the classification of the letter area and the picture area of the entire binarized mixed image is finished. Otherwise, the control unit 10 performs step (32). In step (32), the controller 10 covers the 7 × 7 window on the binarized mixed image of the area to distinguish between the letter area and the picture area. When the 7 × 7 window is written on the binarized mixed image, the controller 10 performs 4-adjacent pixel grouping.

Here, the 4-adjacent pixel grouping will be described in detail. The controller 10 performs 4-adjacent pixel grouping as shown in Table 1 in the binarized mixed image in the 7 × 7 window.

UC RC C LC DC

The controller 10 stores the positions of the pixels in the database when all of the pixels UC, DC, LC, RC on the top, bottom, left, and right sides of the center pixel C of Table 1 have the same value as the center pixel. Then, the pixel at the position is regarded as the center pixel, and the values of all the pixels in the top, bottom, left and right are compared with the center pixel.

In particular, in the present invention, a group of black pixels is made in a 7x7 window by setting the value of the center pixel as a value corresponding to the black pixel.

By performing 4-adjacent pixel grouping in this manner, the number of black pixel groups in the 7 × 7 window and the number of black pixels in the maximum black pixel group, that is, the maximum black pixel group area, can be retrieved.

When the 4-adjacent pixel grouping ends, the controller 10 performs step 36. In step 36, the controller 10 searches for the number of black pixel groups in the 7 × 7 window. When the search of the number of black pixel groups ends, the controller 10 performs step 38. In step 38, the controller 10 searches for the maximum black pixel group area. When the search for the maximum black pixel group area ends, the controller 10 performs step 40. In step 40, the controller 10 searches whether the number of black pixel groups is greater than the threshold 4 and less than the threshold 3 of the maximum black pixel group area. At this time, the controller 10 performs step (42) if the number of black pixel groups is greater than the threshold value 3 and less than the threshold value of the maximum black pixel group area (3). In step 42, the controller 10 determines the binarized mixed image area corresponding to the corresponding 7 × 7 window as the picture area. When the determination is finished, the controller 10 performs the step (30). In operation 44, the controller 10 determines the binarized image area corresponding to the corresponding 7 × 7 window as the font area. When the determination is finished, the controller 10 performs the step (30).

As described above, the present invention distinguishes whether a certain portion of the binarized image is a text area or a picture area. Accordingly, upon enlargement or reduction of the binarized mixed image, it is possible to enlarge or reduce according to the image.

As described above, the present invention distinguishes whether a certain portion of the binarized image is a text area or a picture area as described above. Accordingly, when enlarged or reduced the binarized mixed image is enlarged or reduced according to the image, there is an advantage that can be enlarged and reduced the binarized mixed image without large distortion.

Claims (3)

A method of distinguishing a text area from a picture area in a binarized mixed image, the method comprising: performing a neighboring pixel grouping step of performing adjacent pixel grouping using a value of a black pixel as a center pixel value in a region of a predetermined size of the binarized mixed image; After the adjacent pixel grouping, the black pixel group number and the maximum black pixel group area searching step for searching the black pixel group number and the maximum black pixel group area, and the black pixel group number are larger than one specific threshold, the maximum black pixel group And dividing the picture area into the picture area when the area is smaller than another specific threshold value, and dividing the picture area into the picture area. The method of claim 1, wherein the grouping of adjacent pixels comprises covering a 7 × 7 window in an area of a binarized mixed image for dividing a 7 × 7 window into a text image or a picture image, and then using the black pixel value as the center pixel value. Search if the pixels have the same value as the center pixel, store the position of the pixel with the same value, and again compare the values of all the pixels in the top, bottom, left and right with the center pixel value as the center pixel. A method of distinguishing a text area from a picture area in a binarized mixed image characterized by grouping. 2. The method of claim 1, wherein the one specific threshold is four and the other specific threshold is three. 3.