KR100193840B1 - Video arbitrary magnification conversion method - Google Patents

Abstract

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

A method of converting an image to an arbitrary magnification in a digital image processing system.

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

By dividing the binarized image received from the facsimile into a picture or text area and selectively performing halftone imaging, it provides a method of facilitating enlargement or reduction of the image and without deterioration of image quality.

All. Summary of Solution of the Invention

In this image random magnification conversion method, the binarized image received from the facsimile is divided into a picture area and a letter area, and then processed separately. That is, halftone imaging, that is, binary image, is converted to gray level data by low-pass filtering on the image area, and the gray level data is converted to an arbitrary magnification, and the scaled data is binarized again. It is characterized in that the magnification is converted as it is.

la. Important uses of the invention

It is used to print a small magnification or reduction of images without loss of quality in facsimile machines and copiers.

Description

Video arbitrary magnification conversion method

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for converting an image to an arbitrary magnification in a digital image processing system, and more particularly to a method of magnification or reduction of an image by a few times without deterioration of image quality.

In general, an image processing system sometimes needs to convert an input image to an arbitrary magnification and output it. In the case of facsimile, the communication standard set by CCITT is 203dpi (dot per inch) and the transmission is 203dpi, whereas the resolution of printer is rising to 300dpi or 600dpi. ) Conversion became necessary. In other words, if a 300dp printer is used, the image data received from the facsimile is 203 dpi, so the 203 dpi image data must be converted to 300 dpi. Even in the case of a printer, since the result of printing arbitrary image data desired by the user on the paper varies, the necessity for magnification and reduction of the magnification in the inch and centimeter systems is increasing.

There are two known algorithms for zooming in or out on binarized images. The first is the expansion or reduction of an integer multiple, which simply increases or decreases the number of pixels forming the image by repeatedly outputting or missing the same pixel, and then trims the edges of the image. Since the binary image received by the facsimile consists of two values, black and white, for example, assume that the image is enlarged twice, and the portion of the image to be enlarged is composed of 'white and white' pixels. Assume that the expansion result is 'white and white and black and white'. The second is to simply binarize the binarized image after interpolation without pre-prossing.

Like the first method, simply outputting or omitting the same pixel repeatedly cannot be enlarged or reduced by a few times. In fact, when converting the above-described 203dpi image data to 300dpi, the magnification should be increased by about 0.48 times. However, the repetitive output or omission of one pixel unit is impossible to implement. In the second method, the image quality can be significantly different from the original image. The halftone image input using the scanner is not a problem because of the large amount of data, but since the binarized image received from the facsimile has already lost a lot of information, the magnification of the halftone image has a fatal effect on the image quality. In other words, when binarizing an image in a facsimile or digital copier, the text area is simply binarized, but the picture area is dithered or error diffusion, so it is simply enlarged or reduced. In this case, the original image may result in the overlapping or connecting of the separated points or lines with each other, and thus the image quality may be greatly deteriorated compared to the original image.

Accordingly, an object of the present invention is to provide a method of facilitating enlargement or reduction of an image and without deterioration of image quality by subjecting the binarized image received from the facsimile to a halftone image by selectively classifying the binarized image. .

1 is a block diagram of a facsimile to which the present invention is applied

2 is an enlarged view of a halftone image and a portion thereof according to an embodiment of the present invention;

3 is a diagram illustrating a halftone image random magnification conversion process according to an embodiment of the present invention;

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. First, in adding reference numerals to the components of each drawing, it should be noted that the same reference numerals have the same reference numerals as much as possible even if displayed on different drawings. In addition, in the following description, many specific details such as components of specific circuits are shown, which are provided to help a more general understanding of the present invention, and the present invention may be practiced without these specific details. It will be obvious to those skilled in the art. In the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

1 is a block diagram of a facsimile to which the present invention is applied. The control unit 111 controls the overall operation of the transmission mode, the reception mode, and the copy mode of the fax according to the set program. In addition, the control unit 111 may include a program memory and a data memory or may be externally connected as shown by reference numeral 119. The program of the present invention is stored in the program memory, and the data generated during the execution of the program is temporarily stored in the data memory. The OPE 112 includes a key input unit 220 and a display unit 210. The key input unit 220 of the OPE 112 generates data of a key pressed by the user to designate each mode and perform an operation of the designated mode, and output the data to the controller 111. In addition, the display unit 210 of the OPE 112 performs a function of inputting and displaying display data indicating an operation state of the system when the controller 111 performs each mode. The sensor unit 113 detects the presence or absence of an input document and print paper, and outputs the detected state signal to the control unit 111. The scanner 114 transmits the data, which the CCD (not shown) reads image data from the original, photoelectrically converts it, and outputs the data to the controller 111. The image processor 115 encodes and decodes the image data output from the scanner 114 and the image data output from the modem 117. That is, the image data input from the scanner 104 is divided into a background and a text or a half-tone image data is processed so that the receiving side can receive the image almost identical to the actual image. do. The print unit 116 prints the image data input from the image processing unit 115 on a print sheet when the receiving mode and the copy mode are controlled under the control of the control unit 111. The modem 117 is controlled by the control unit 111 and performs analog modulation on the data output from the control unit 111 and digitally demodulates the signal received through the transmission line. That is, in the transmission mode, the image data output from the image processing unit 115 is modulated and transmitted in accordance with the transmission type-the standard of the facsimile, and in the reception mode, the modem 117 processes the encoded image signal input through the transmission line. Demodulate in original form so that you can The network control unit (NCU) 118 forms a transmission / reception path between the telephone line (tip, ring) and the modem 117 under the control of the control unit 111.

According to the present invention, information of data can be artificially increased for facsimile images received by binarization, so that enlargement or reduction of less than 1% is possible. The binarized images received by the facsimile are mostly Bayer dithering or error diffusion binarized images. In this image, the dark part has a lot of black pixels, and the bright part has many white points, and thus its characteristics can be used. In other words, the gray level distribution from the binarized image is forcibly restored to the gray level value, that is, the contrast. The method used to recover this lost information uses a smoothing algorithm called a lowpass filter. According to the low pass filter, a place where many black points have a dark value sequentially darkens, and a place where black points have sparse distributions has a light gray level value. In the present embodiment, the following two conditions are set based on these characteristics, and the experiment is repeated to set threshold values for each. The first condition is that the black spots do not appear in a row in the algorithm of error diffusion or dithering. In this embodiment, the threshold (first threshold) is set to five. The second condition is that the proportion of the distribution where the black dots are adjacent in the horizontal and vertical directions in the direction of the pixel is high in the area where the text appears. In this embodiment, the threshold (second threshold) is set to five.

2 is a diagram illustrating a 7 × 7 masking state of a facsimile received image and a portion thereof according to an embodiment of the present invention. (2a) is an image received from the facsimile, (2b) shows a state in which the 7 × 7 mask is applied to a part C of the picture portion A in (2a), (2c) is a character in (2b) The 7x7 mask is shown with respect to the part D of the part B. FIG.

3 is a diagram illustrating a binarized image random magnification conversion process according to an embodiment of the present invention. In this embodiment, a case where the binarized image received by the facsimile is converted to an arbitrary magnification will be described as an example. It is also assumed that the first and second thresholds are five, respectively. In step 3a, the control unit 111 receives the binarized image transmitted from the other party through the telephone line, the NCU 118, and the modem 117. In step 3b, binary values (black and white) of three lines are read from the binarized image. Move the 7 × 7 mask and check whether the area is a picture or a text. Referring to (2c) of FIG. 2, a mask is covered around an arbitrary pixel Q (4,4) and the characteristics of the surrounding pixels are examined. At this time, nine consecutive points are formed around the pixels Q (4,4). The number of consecutive points 9 is greater than the first threshold value 5 according to the first condition described above. In addition, since the nine points are connected in the vertical and horizontal directions, this value is also larger than the second threshold value 5 according to the second condition described above. Is recognized. If the check result is found in the writing region in step 3c, the process proceeds to step 3d to generate the font position information for the pixel Q (4,4). That is, the coordinates of the point are stored as text points. On the other hand, if it is determined in step 3c that the area is not a letter part, the process proceeds to step 3f and checks whether it is a picture part. Referring to (2b) of FIG. 2, when the distribution of black points is examined centering on an arbitrary pixel P (4,4), the pixel P (4,4) has one continuous black point. Therefore, the pixel P (4,4) is smaller than the first threshold value 5 and is recognized as a point (picture portion) having characteristics of a picture. As such, when the check result is found to be a picture part, in step 3g, picture position information of the pixel P (4, 4) is generated. That is, the coordinates of the points are stored as picture points. The location information of all the processed pixels (images) is stored in a specific area of the memory, and the picture area and the text area are classified and processed based on the location information.

Specifically, in the case of the text area, the process proceeds to the step 3e to enlarge or reduce the prime number (less than 1%) by the basic algorithm such as expanding by the first interpolation method or reducing by the resampling method. And if necessary, edge enhancement technique is performed. On the other hand, in the case of a picture area, the process proceeds to step 3h and inversely transforms (restores) the halftone image data, that is, the gray level data before binarization, through low pass filtering (averaging). At this point, even though the average of many black spots is averaged, the black gray level value is obtained and the opposite area has the opposite gray level value. Subsequently, in step 3i, the gray level data is enlarged or reduced by a small number of times (less than 1%) using basic algorithms such as first-order interpolation or resampling. In step 3j, the decimal magnified or reduced data is binarized again by dithering or error spreading. In step 3k, the control unit 111 reads the stored data from the memory, transfers the stored data to the print unit 116, and outputs it to the paper.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications may be made without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the scope of the following claims, but also by the equivalents of the claims.

Once the binarized image is converted to gray level to have more information (multiple levels of brightness), it can be enlarged or reduced, and then binarized to fit the printer with higher resolution. You can get much smoother and more sophisticated images. In addition, by converting the binarized image into a virtual gray level to enlarge or reduce it, it is easy to enlarge or reduce the number of times.

Claims (4)

In the method for converting a binarized image to an arbitrary magnification in a digital image processing system, The binarized image is divided into a picture area and a letter area, and the picture area is converted into a halftone image to convert the magnification and then binarized again, and the letter area is converted into a magnification to generate a new binarized image. Way. In the method for converting a binarized image to an arbitrary magnification in a digital image processing system, Dividing the binarized image into a picture area and a letter area, and storing location information thereof; The binary data corresponding to the picture area is converted to a halftone image having gray level data by low pass filtering according to the position information, and then scaled with the converted gray level data, and then binarized again. Binary data is a process of generating a new binarized image by scaling conversion as it is. 3. The method of claim 2, wherein the arbitrary magnification is a magnification for scaling up or down a fraction of 1%. In the method of magnification or reduction of the binarized image received from the facsimile, In the binarized image, when moving a mask of a predetermined unit, the black area is judged as a picture area when there are more than a first threshold value in series with respect to each point based on the point, and the black point is adjacent in the horizontal and vertical directions based on the point. When the ratio of the distribution is greater than or equal to the second threshold, determining the text area and storing the location information; According to the position information, the binary data corresponding to the picture area is low-pass filtered and converted into a halftone image having gray level data, and then scaled up or down by a small number of times with the converted gray level data, and then binarized again. The binary data corresponding to the letter area is characterized in that the process of generating a new binarized image by expanding or reducing the number of times as it is.