JPS61247165A - Method for dither picture coding - Google Patents

Abstract

PURPOSE:To improve the coding efficiency by rearranging a binary-coded picture at the same threshold value in a matrix so that picture elements binary-coded by the same threshold value in the matrix approach together. CONSTITUTION:An intermediate signal inputted from an input section (not shown) is stored in an intermediate tone signal memory 1, the result is divided into blocks comprising plural picture elements and each picture element is compared with a threshold value at a comparator 2. The threshold value differs from each picture element and is stored in a threshold value storage memory 3. The signal binary-coded by the comparator 2 is stored in a binary signal memory 5 via a memory address controller 4 and the binary-coded picture elements by using the same threshold value are rearranged so as to approach together. Then the result is coded by a binary signal coder 6 and sent. Thus,the coding efficiency is improved.

Description

[Detailed Description of the Invention] The present invention relates to a dithered image redundancy reduction encoding method in which a dithered image is applied to an existing binary encoding method. The present invention relates to a dither image encoding method that makes it easy to encode a dithered image.

(Prior art) When transmitting an image containing halftones to a device capable of only binary display, such as a facsimile, the image is divided into blocks each consisting of a plurality of pixels, and each pixel in the block is divided into two blocks with different threshold values.
A method is known in which the digital signal is converted into a value and the binary code is transmitted instead of the halftone signal. .

(Problem to be solved by the invention) However, in order to display a high-quality halftone image on the receiving side,
The binarization code is 1 for every 1 to several pixels. The result is a complex signal that changes rapidly between o (or white and black), so the conventional binary signal encoding method (modified) which is based on the transmission of one document, one figure, etc. (such as the human method or modified read method) is Even if applied, there was a drawback that the encoding efficiency did not improve.

On the other hand, a signal that is binarized using one of the relatively widely used dither threshold matrices called Bayer type has a pixel arrangement of 1, 0 (or white, black) similar to a checkerboard pattern, so all pixels are connected to each other. There is a method of separating the pixels into two checkerboard grid pixel groups that interpolate with each other so that 1 or 0 (white or black) is likely to continue. However, although this method is effective to some extent for the Bayer type, it has the drawback that it is almost completely ineffective for other types of threshold matrices.

(Means for Solving the Problems) In order to solve these drawbacks, the present invention solves these drawbacks by dividing a binarized image between pixels that have been binarized using the same threshold in a matrix. The arrangement is arranged so that they are closer to each other.

(Function) By doing this, even if the shape of the threshold matrix changes, 1 or O (white or black) will continue, and the existing 2
It becomes easier to apply the value signal encoding method.

(Example) FIG. 1 is an example of a dither threshold matrix applied to a block consisting of four pixels, and ■ and T4 are different thresholds.

FIG. 2 shows pixels that have been binarized using the matrix shown in FIG. 1, and All to A4+ are pixels that have been binarized using thresholds 11 to 4 in FIG. 1, respectively. Further, i indicates the order in which the images are binarized when dither processing is performed starting from the upper left of the image.

FIG. 3 shows an example of pixel rearrangement according to the present invention, in which pixels binarized using the same threshold are arranged in the same area. (A, ~A4 are the same as in Figure 2) Therefore, for example, if 80 is black and Azi is white, in the conventional Figure 2, white and black appear alternately, and the existing encoding method is extremely inefficient. However, in FIG. 3, black and white are continuous, and a long continuous length of black or white can be encoded with a short code, increasing efficiency.

Figure 4 shows an example of rearranging pixels that have been binarized using the same threshold to bring them closer to a set of pixels that have been binarized using a threshold that is close to that threshold, in order to further improve efficiency. It is.

For example, if the threshold values have a magnitude relationship of T, >T, >T, >T in FIG. 1, if a set of A41 pixels is placed in the area where Azi pixels are gathered in FIG. 2, T, , T.

If the values are close to each other, at+ and A41 often have the same value, so the length of consecutive black or white pixels becomes even longer, increasing the encoding efficiency.6 Also, the two-dimensional correlation between white and black In the encoding method using A41 in Fig. 4,
It is sufficient to place the group below the group A1.

FIG. 5 is a further modification of the above, and is an example in which groups of pixels binarized with each threshold are arranged while maintaining the magnitude relationship of the thresholds. The magnitude relationship of the threshold values is for the case of □〉〉〉〉〉〉〉〉2〉.

In this case, the probability that white and black pixels are consecutive is A.

It becomes high not only between the A4i groups, but also between the A41tAii groups, and between the A□ and A1 groups, and the coding efficiency improves.

In particular, in FIG. 5, the pixel groups are rearranged in order of magnitude of the threshold values, but they may be rearranged regardless of the magnitude of the threshold values.

In that case, the encoding efficiency will be slightly lower than in the case of FIG. 5, but the processing becomes simpler because it is not necessary to consider the magnitude of the threshold value.

Also, the method of arranging pixel groups one after another in a horizontal row as shown in FIG. 5 is compared to the method of arranging pixel groups two-dimensionally as shown in FIGS. 3 and 4.

The order in which pixels are binarized using the dither method and the order in which they are input to the binarization encoder are relatively close.

At the top of the rearranged image, the pixels with the lowest binarization order, that is, the pixels with the smallest number of j in Aii, are clustered.

Therefore, it is effective when pursuing real-time performance.

FIG. 6 is a block diagram of an embodiment showing the configuration of the present invention, in which 1 is a halftone signal memory, 2 is a comparator, 3 is a threshold storage memory, 4 is a memory address controller, and 5 is a binarization signal. Memory 6 is a binary encoder.

The halftone signal input from the input section is once stored in a memory, and then compared with a threshold value and converted into a binary signal.

The memory address controller determines the threshold value used in the comparison, controls the memory address so that pixels binarized with the same threshold value are located close to each other, and stores the binarized code in the memory.

Thereafter, the binary code is encoded and transmitted by an encoder.

FIG. 7 is a diagram showing the configuration of another embodiment of the present invention, 1
Components 6 to 6 are the same as in FIG.

The difference with Fig. 6 is that in Fig. 6, pixels are rearranged at the same time during binarization, whereas in Fig. 7, after the binary signal is stored in memory, the binarized code is When transferring to the device 1
This is to rearrange the pixels.

As explained above, by using the present invention, existing binary encoding methods can be applied very efficiently.

For example, when using a Bayer-type matrix with a size of 4 x 4 pixels, if it is encoded as is using the modified read method, which is the standard encoding method for facsimile, approximately 2 bi
The code amount is t/pixel, but if the present invention is used, the code amount is 171
It can be encoded with approximately 0.18 bits/pixel of 0 or less.

Furthermore, even in matrices other than the Bayer type, the encoding efficiency hardly changes because the positions of the pixel groups binarized using the same threshold value in FIG. 3 only move macroscopically.

In the description, the size of the matrix is set to a 2×2 square area and the number of thresholds is set to 4, but these are just examples, and the shape and number may be changed.

In addition, in the example, for explanation purposes, we considered a case in which the entire screen is rearranged at the same time, but due to the amount of processing and memory, there is no problem and the present invention can be applied to every several divisions of one screen, and the effect will be more effective. is almost unchanged.

[Brief explanation of drawings]

Figure 1 is an example of a dither threshold matrix applied to a block consisting of four pixels, Figure 2 is an image binarized by the dither method, Figure 3 is an example of pixel rearrangement according to the present invention, and Figure 4 is an example of a dither threshold matrix applied to a block consisting of four pixels.
The figure shows an example of pixel rearrangement in which pixel sets with similar threshold values are brought closer together according to the present invention, and Figure 5 is a modified example of Figure 4, which is a group of pixels that are binarized at each threshold value while maintaining the magnitude relationship of the threshold values. FIG. 6 is a block diagram of one embodiment showing the configuration of the present invention, and FIG. 7 is a diagram showing the configuration of another embodiment of the present invention. 1...Halftone signal memory, 2...Comparator, 3.
...Threshold value storage memory, 4...Memory address controller, 5...Binarization code memory, 6...Binarization code encoder, T1 to T4...Threshold value, All""A4i.
・Pixels that were binarized from 11 to 4 respectively, i ・・
- Binarized order. Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7

Claims (3)

[Claims] (1) In a system that binarizes a halftone image signal using a dither threshold matrix and transmits it, the pixel positions of the binarized signal are rearranged so that pixels binarized with the same threshold are close to each other. 1. A dithered image encoding method, characterized in that an image signal that has been processed and rearranged is encoded using an existing binary signal encoding method. (2) Rearrange the pixel positions so that the set of pixels binarized with a threshold value close to that threshold value approaches the set of pixels binarized with the same threshold value, and then The dither image encoding method according to claim 1, characterized in that encoding is performed using a signal encoding method. (3) The pixel positions are rearranged so that a pixel set of pixels binarized with the same threshold is connected one after another in a fixed direction, and then encoded using the existing binary signal encoding method. Claims (1) or (2) characterized in that:
The dither image encoding method described in section 2).