JPS60201479A - Compression system of picture data - Google Patents

Abstract

PURPOSE:To obtain a compression system of piclture data with simple operation and high efficiency by collating a data of effective component only where the picture data compression of coding in the unit of division compared with a specified pattern of a specified storage area is <=64 bits, transmitting the compressed picture data and transmitting a raw data as it is as to the picture data without compressing efficiency. CONSTITUTION:A frame division section 3 has a function dividing an inputted picture data having horizontal p-bit and vertical q-bit into an optional processing division units m X n bit. A collating section 4 accesses sequentially the dot pattern constitution at each received division unit in the order of code and when the pattern is coincident with the specified pattern, the section 4 transmits the code to a data amount comparator section 5. The section 5 counts the data amount of coded display per division unit and informs it to a control section 1 when the amount exceeds 64 bits, the control section 1 allows the collation section 4 to stop the coding of the specified pattern in the said division unit and transmit the 64 bit data while being not converted. In transmitting the non-conversion data, a control signal representing that the 64-bit data is a raw data is added.

Description

DETAILED DESCRIPTION OF THE INVENTION (a) Technical Field of the Invention The present invention relates to an image processing system, and more particularly to an image signal compression method.

(b) Background of technology Although human beings use their senses to understand information about a certain object, it is a well-known fact that the visual process is the best way to understand the content. A data input/output means that takes full advantage of the real-time processing function of a data processing system, especially a cathode ray tube (C
Display output using a display such as RT) or means for printing and/or drawing on a hard copy using a printing means are widely used.

The disadvantage of data processing using images is that the amount of information involved is extremely large, so storage or transmission of the data requires large-capacity storage means and high-speed transmission means, which increases the size of the equipment and increases the cost. . In recent years, with the development of semiconductor technology, especially integration technology, and high-density technology in magnetic storage, storage means with low cost per focus and high-speed processing technology have gradually been realized, and image processing costs are gradually becoming lower. With the spread of data processing technology, higher speed processing at lower costs is expected.

Conventionally, image processing involves converting a real image document into an electrical signal (analog data) by optical scanning, and converting the signal from analog to digital (analog data) to facilitate data processing. (A/D) conversion, digitization, and processing. The stored or transmitted image information is restored to the original real image original image by scanning and reproducing, and is outputted by the display/printing means as described above. These accumulation #
r/ In order to compress the amount of image data as much as possible during the transmission process, reduce the burden of storage/transmission, and realize efficient image processing, image data has traditionally been processed according to certain conventions, such as those of the International Telegraph and Telephone Advisory Committee. (CCITT) T・4
Image data compression processing, such as one-dimensional and two-dimensional No. 49 encoding in Group 3 of high-speed digital facsimile (FAX) defined in 2003, is also widely put into practical use.

That is, the transmitting side of the image processing system c' is equipped with an image data compression/encoding function and transmits the compressed image data, and the receiving side executes the halo/decoding process of the compressed image data.

(C) Conventional technology and problems When conventionally performing image signal compression/encoding processing 3- As seen in T.4 above, compression processing is performed by focusing on changes in the image signal, so white Alternatively, when black data is continuous, the amount of compression is large and efficient; however, if there are many changes from white to black or from black to white, the amount of compression becomes small and the compression efficiency decreases. This also applies vertically,
If the change in the repeating white/black pattern in the horizontal set meal line is repeated in the next scan line, the next scan line will require only a small amount of data and the amount of compression will be large. Therefore, when there is a spot on the white background or a missing part on the black background of the image original data, there is a drawback that the data 11 increases because the data frequently changes in both the horizontal and vertical directions as the white/black changes.

(d) Purpose of the Invention The purpose of the present invention is to eliminate all of the above-mentioned drawbacks by dividing image original data, which is normally rectangular, at equal pitches into X and Y grids of integer multiples of image dots when compressing/relativizing image signals. Each area is defined as a total compression unit, and a specified pattern that can be superimposed once the number of times is registered in the unit, and the image original data to be subjected to pressure lines is compared with the specified pattern for each of the 41 units to form the specified pattern. When a matching image unit is obtained, the image unit is converted into a code of a specified pattern, and for image parts where there are many changes in black and white and the compression effect cannot be obtained, the image data is sent out using raw don't data. The objective is to provide a highly effective compression method for image data.

(e) Structure of the Invention The purpose is to compress/compress image data according to a certain hand IF.
In an image processing system that encodes data, stores it, transmits it, and reproduces and restores it to image data as necessary,
The image data compression function applies to one piece of image data as a processing unit.
Means for dividing the frame into units of mxn dots Storage means for registering the dot pattern configuration of the number of types in one division unit as a prescribed pattern in correspondence with the identification code For each division unit sent by the one division means A matching means for successively comparing the dot patterns included in the unit from the higher-order pattern to the lower-order pattern in the storage means and sending out a matching specified pattern code; The control unit in the image data compression function applies voltage to the entire input image data division means, sequentially inputs it to the collation means for each division unit, and stores the regulations stored in the storage means. The pattern code is sent out by comparing it with the pattern, and the comparing means totals the pattern code data amount in the division unit, and for the division unit where the pattern code data amount ≦m><n, a unit identification flag is added to the first code. The pattern code data bar > m x n
When detected, the processing of the division unit after the collation means is stopped, and the dot data sending signal is added to mxn.
This can be achieved by providing a compression method for image data that has all the characteristics of transmitting it as raw dot data.

(f) Embodiment of the Invention An embodiment of the invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of an image data compression method according to an embodiment of the present invention, FIG. 2 is an example of division in an image data frame IC, and FIG. 3 is a diagram of a prescribed pattern within a division unit and its pattern code. Correspondence example diagram, FIG. 4 is a diagram showing an example of the configuration of the prescribed pattern code, FIG. 5 (a),
(b) is an example of standard pattern matching at 1 division rate.
6 is an example of compressed/uncompressed data for the division units shown in FIGS. 5(a) and 5(b). In the figure, 1 is a control section, 2 is a storage section, 3 is a frame division section, 4 is a collation section, 5 is a data amount comparison section, and 20 is a prescribed pattern storage area. The III control unit 1 is composed of, for example, a microprocessor (MPU), and controls the other components of the image data compression function according to the storage unit 2 (not shown) or its own built-in control program and control unit data to process the image data. Perform compression processing.

The storage unit 2 has a function for storing control programs, control data, and other data necessary for compressing image data, such as prescribed patterns to be described later. Full frame division capital 3 is input image data horizontally 7-p bits, vertically q bits, arbitrary processing division unit mX
It has a function to divide into 11 bits. For example, if an A4 paper has 11728 bits in the horizontal direction and m=8, the horizontal direction will be divided into 216 bits. For convenience of explanation, m
An example is shown in which f8 is set to 8 bits. Here, the prescribed pattern in division units will be explained. 103 types of prescribed patterns are set for the 8×8 configuration of the division unit. The first group A is code O~3 in Figure 3.
and 13 types of 4 x 4 dot configurations corresponding to numbers 4 to 12,
Second group B is code 13-28 and 29-37
2Bffi of 2×2 dot configuration corresponding to , 64 types of 1 dot configuration corresponding to codes 38 to 101 in group C of i3, and 1 type corresponding to all white (black) code 102 are stored in the specified pattern storage area. It shall be assumed that The collation unit 4 accesses the dot pattern structure of each division unit received from the frame division unit 3 in code order [0 to 102] one by one, and when they match, sends the code to the data amount comparison unit 5. Here, if there is no restriction on the amount of data regarding the expression of the dot pattern configuration 8- in the division unit, code 38~
Since 101 is prepared, coded display is possible for division units with any pattern, but coded display displays one specified pattern by 8 pins XC shown in Figure 4, so coded display is possible. If the display exceeds about 8 bits per code, the compression effect cannot be obtained. That is, in the example shown in FIG. 5(a), 24 pits and 6
Since it is 4 bits, there is an effect of 411 outside compression, but as shown in Fig. 5 (
In the example b), the code ratio display σ month 04 bit) is 64 pi nod, which is rather disadvantageous. The data amount comparison section 5 counts the data amount of the coded display of the division unit 9, and when it exceeds 64 bits t-, it notifies the control section 1 to that effect, and the control section 1 causes the collation section 4 to All 14 encoding operations of the specified pattern in units are stopped and the 64-dot data is controlled to be sent out without being converted.

When unconverted data is sent, a control signal indicating that the 64-dot data is raw data is added.

In addition, if the coded display according to the specified pattern is 64 dosits or less, it is sent as a compressed data using i' j - codes, but in this case, the data following the first bit of the division unit is the first bit of one division unit. A division unit identification flag, for example 0, is added to indicate that this is the case, and the data is sent out as shown in FIG. The remaining 7 bits are code 0.0.0 as shown in FIG. 5(a). The prescribed pattern according to ■ is shown. Therefore, the first 8 bits are 0.
The following 7 bits become 0001100 indicating ■. At the beginning of the next 8 bits, 1 is set in the identification flag of the division unit, indicating that this data has been generated overlappingly in the division unit with O, and on the receiving and reproducing side, the coded data following 1 is set. will be overlapped and output in the same division unit. Since this embodiment is configured as described above,
The collation unit 4 is operated to send out compressed image data only for valid portions of image data for which compression of the image data by coding in division units is 64 bits or less compared with the specified pattern in the specified pattern storage area 20, resulting in inefficient compression. Since the image data is sent out as raw data, an efficient image data compression method with a simple configuration and operation of the image data compression function can be obtained. Although a positive image of black data on a white background has been described above, it goes without saying that a negative image of its inverse image can also be realized in the same manner.

(g) As described in detail, according to the present invention, an image frame is divided into equal pitch unit areas as processing units, and image data for each unit is compared with a predetermined pattern of counts accumulated in advance. All of the image data that matches the specified pattern is encoded and sent out, and when the number of coded data reaches 11 and there are 14 division units where no compression effect can be obtained, the raw data is sent and filtered without using compressed data. 0, which provides a highly effective image data compression method with a simple configuration.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of an image data compression method according to an embodiment of the present invention, Fig. 2 shows how many times the image data frame is divided into VC, L-, and Fig. 3 shows the prescribed pattern within the division unit and its pattern. Correspondence example diagram with code, 4th
The figure shows the number of times the 11-configuration of the prescribed pattern code, Figure 5 (a), (b) shows the number of times the prescribed pattern is matched in one division unit, and Figure 6 shows Figure 5 (a),
How many times is the compressed/uncompressed data corresponding to the division unit in (b)? 12- Figure 1 Dormitory Figure 2

Claims (1)

[Claims] In an image processing system that converts image data into compressed/encoded data according to a certain procedure, performs storage and transmission processing, and reproduces and restores it as necessary image data, the image data compression function is an image processing unit. 1 frame of data im
Is it made up of Xn dot units? Means for dividing into Jl number of units Storage means for registering a plurality of types of dot pattern configurations in two division units as prescribed patterns corresponding to the identification code One dot pattern included in each division unit sent by the division means A matching means successively compares the upper defined pattern to the lower defined pattern in the storage means and sends out a matching defined pattern code, and the amount of data in the pattern code sent out by the matching means is aggregated for each division unit and mXn The control unit in the image data compression function marks the input image data on the dividing means.
In addition, the pattern code is sent out by sequentially inputting it into the matching means for each division unit and comparing it with the prescribed pattern stored in the storage means, and the comparison means is made to total up the amount of pattern code data in the division unit, thereby determining the amount of pattern code data. ≦mxn
For the divided unit, a unit identification flag is added to the leading code thereof, and data according to the pattern code is sent out, and when a pattern code data amount>mXn is detected, processing of the divided unit after the collation means is stopped, An image data compression method characterized by adding a dot data transmission signal and transmitting it as mxn raw dot data.