JP2601169B2 - Image data compression coding system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system for compressing and encoding image data, and more particularly to a method for compressing and encoding image data by paying attention to the relationship between adjacent pixels, thereby efficiently performing image processing such as image transmission and storage. For compressing and encoding image data.

[0002]

2. Description of the Related Art Conventionally, as this kind of image data compression encoding system, there is one disclosed in, for example, Japanese Patent Application Laid-Open No. 60-185468. In this image data compression encoding system, a compression code is generated from a number in which the same difference as the difference value for each pixel continues. First, the following Table 1 is used to indicate which component of red, green, and blue has a difference.
The differential mode code shown in FIG.

[0003]

[Table 1]

Subsequently, only the difference value of the component having a difference is represented by Wyl as shown in the following Table 2 in the order of red, green and blue.
Compression encoding was performed using a variable-length code such as an e-code.

[0005]

[Table 2]

[0006] Here, S in Table 2 is a sign bit indicating the sign of the difference value, which is "0" when the difference value is positive and "1" when the difference value is negative.

[0007]

In the conventional image data compression encoding system described above, when decoding is performed by software using a microprocessor that controls the entire system, the compressed code bits are cut out one by one.
It is determined how many bits the difference value to be obtained is composed of, and thereafter, data is cut out from the compressed code bits by the number of bits of the difference value, and one difference value is obtained. Moreover, since the difference value data exists for the size of the image for each of the three color components, the number of processing steps in the microprocessor becomes enormous, the access to the memory becomes complicated, and the processing time is increased. was there.

It is an object of the present invention to provide an image data compression encoding system capable of processing data at high speed.

[0009]

SUMMARY OF THE INVENTION The present invention relates to an image data compression encoding system for compressing and encoding image data composed of three components of Y (luminance signal), Cr and Cb (color difference signal). To a plurality of blocks, a means for calculating a difference value of the Y (luminance signal) between adjacent pixels in each block unit,
Information on whether r and Cb (color difference signals) have the same value as the adjacent block and the difference value of the Y (luminance signal)
classified into bit identifiers, and Cr and C in the block.
Which of the difference between the value of the b (color difference signal) component and the difference value of the Y (luminance signal) component for the number of pixels in the block is included in the code string and which value is not included in the code string, Means for classifying the value into a plurality of coding patterns for designating how many bits the value is to be included in, and an integer of n bits (n represents the bit precision of the Cr and Cb) for each coding pattern Means for creating a code string represented by a double.

[0010]

According to the present invention, by performing coding by dividing into a plurality of cases by using an identifier, it is possible to perform processing with a small number of steps according to each case at the time of decoding, and to shorten the decoding processing time in the microprocessor. It is possible to do.

[0011]

Next, the present invention will be described in detail with reference to the drawings.

Image data to be encoded is represented in a color space of Y (luminance signal), Cr, Cb (color difference signal), and downsampling of the image is performed by Y: Cr: Cb.
= 4: 1: 1, image bit precision is Y6 bit, Cr
An example in which the compression data is represented by 4 bits and Cb 4 bits and the code length of the compressed data is fixed to 4 bits will be described with reference to the drawings.

FIG. 1 is a block diagram showing an apparatus of an image compression system according to an embodiment of the present invention. In FIG. 1, a central processing unit (CPU) 1 operates by executing a program stored in a control ROM 2 using a control RAM 3 as a work area.

The file device 4 contains original image data to be compressed. The CPU 1 accesses the file device 4 to read data and perform a compression process. The compressed image data is written to the file device 4. When displaying the compressed data, the CPU 1 operates the file device 4
Is accessed, the compressed image data is read and subjected to decompression processing, and the decompressed image data is written to the frame buffer memory 8 to display an image on the display device 9.

FIG. 2 shows an example of a data structure for performing the above processing at high speed. However, in this case, each code data is 4
It consists of a bit identifier followed by data in 4-bit units.

FIG. 2A is a diagram showing the relationship between the pixel position in the block and the difference values (Y0 to Y3).

The difference Y0 is the difference between the lower right pixel of the block one block before and the upper left pixel of the target block to be coded. In the case of the first block of the image (the upper left block of the image), A difference value is calculated assuming that the value of the lower right pixel of the block one block before is 0.

Code data is determined based on the difference value. In a general natural image, the difference between the difference between adjacent pixels and the number thereof is as shown in FIG. There is a property that there are many values near the value 0 and small values near the value. For such data, the compression efficiency is improved by encoding the data of 0 without encoding the data close to 0 with a code having a small number of bits, and encoding the data far from 0 with a code with a large number of bits.

FIG. 2B is a diagram showing an example of a data structure according to the present invention.

The image data to be encoded in the present embodiment is such that the downsampling of the image is performed by Y: Cr: C
Since b = 4: 1: 1, Cr and C are contained in one block.
There are a total of six data, b, Y0, Y1, Y2, and Y3.

In the figure, one square represents a 4-bit code, and there are fourteen types of encoding patterns 201 to 214 classified as 4-bit identifiers.

CrCb is coded with 4 bits of raw data, and the difference values of Y are all 4 bits or less (−8 to +
In the case of 7), encoding is performed with the data configuration shown in 201.
(In this case, the number of bits is 28 bits for four pixels.) If there is even one piece of data of 4 bits or more in the Y difference value, encoding is performed using the data configuration shown in 202. (In this case, the number of bits is 44 bits for four pixels.) When all four Y values are the same (in the case of a difference value of 0), one Y data is represented by 203 and 20.
Encoding is performed using the data configuration shown in FIG. Similarly, if there are coding patterns from 205 to 206 depending on the difference value of Y, and if the values of Cr and Cb are the same as those of the previous block, the coding is performed with the data configuration shown in 208 to 214.

Next, the compression processing will be described with reference to the flowchart of FIG.

When the compression process is started, an original pixel file is read from the file device 4 (step S401), and blocks are cut out (step S402). In the encoding process, the image data is processed every two lines in the direction from the upper left to the lower right of the image in a block unit of 2 pixels vertically and 2 pixels horizontally.
It is checked whether or not the data of CrCb is the same as the previous block in the cut out block (step S403). If the data is not the same as the previous block, the difference calculation processing unit 5 adjoins the Y data. A difference value from the pixel that has been obtained is obtained (step S404). The process branches to the processing unit corresponding to the pattern of the data structure shown in FIG. 2 based on the difference value obtained in this way (step S405), and performs a process of creating a code string for each pattern of the data structure shown in FIG. Thus, code strings 201, to 207 in FIG. 2 are created (step S406), and the procedure goes to the next step S410.

In step S403, if the data of CrCb in the cut block is the same as that of the previous block, the difference calculation processing unit 5 obtains a difference value between adjacent pixels for the Y data ( Step S40
7). The process branches to the processing unit corresponding to the pattern of the data structure shown in FIG. 2 based on the difference value thus obtained (step S408), and performs the process of creating a code string for each pattern of the data structure shown in FIG. And 208 in FIG.
, 214 are created (step S40).
9) Go to the next step S410.

When the processing of steps S402 to S409 is completed for all blocks, the compressed data is written to the file device 4, and the compression processing ends.

Next, the flow of the decompression process in the present invention will be described with reference to FIG.

When the decompression process is started, compressed data is read from the file device 4 (step S501), identifier data is read (step S502), and cases are classified according to the identifier (step S503). Next, in step S504, data corresponding to the number of bits required for each case is read from the compression code, and data 202 and 202 in FIG.
, And 214 (step S50).
4) The decoded data is stored in the frame buffer memory 8 (step S505). These steps S5
02 to S505 are processed in units of one block for all blocks to decode image data (step S50).
5). Note that the code string need not have a constant length.

[0029]

According to the present invention, it is possible to perform processing with a small number of steps according to each case at the time of decoding by performing encoding by dividing into a plurality of cases by an identifier, and to decode image data at high speed. It becomes possible. Also, the present invention
By increasing the difference value 0, the size of the code data can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram showing one embodiment of the present invention.

FIG. 2 is a diagram illustrating an example of a data structure according to an embodiment of the present invention.

FIG. 3 is a histogram of data in one embodiment of the present invention.

FIG. 4 is a flowchart showing a compression process according to an embodiment of the present invention.

FIG. 5 is a flowchart showing a decompression process in one embodiment of the present invention.

[Explanation of symbols]

 Reference Signs List 1 CPU 2 Control ROM 3 Control RAM 4 File device 5 Difference calculation processing unit 6 Encoding processing unit 7 Decoding processing unit 8 Frame buffer memory 9 Display device

Claims (1)

(57) [Claims] 1. An image data compression encoding system for compressing and encoding image data composed of three components of Y (luminance signal), Cr, and Cb (color difference signal), means for dividing an image into a plurality of blocks. Means for calculating a difference value of the Y (luminance signal) between pixels adjacent to each other in the block unit, and information on whether the Cr and Cb (color difference signals) have the same value as an adjacent block. And a difference value between the Cr and Cb (color difference signal) components in the block and the Y (luminance signal) components for the number of pixels in the block. A means for classifying into a plurality of coding patterns for specifying which value is included in the code string and which value is not included in the code string, and when including, which bit is represented by the number of bits. , Each encoding N for each pattern (where n is the C
means for creating a code string represented by an integer multiple of bits (representing the bit precision of r and Cb).