JPH06164944A - Compression method for picture data, its equipment and decoding method of picture data and its device - Google Patents

Abstract

PURPOSE:To reduce the processing time without deterioration in quality of a decoded picture by forming a compressor of picture data represented in a binary signal with a high-order bit compression data generating means, a low- order bit compression data generating means and a means generating compression picture data based on the data. CONSTITUTION:The picture data compressor is made tap of a high-order bit compression data generating means 11, a low-order bit compression data generating means 12, and a compression picture data generating means 13. The means 11 compresses run length of a signal of a high-order bit section of plural picture elements for each bit string to generate highorder bit compression data and the means 12 generates low-order bit compression data by plural adjacent, picture elements and the means 13 generates compression picture data based on the compression data. Furthermore, the decoder is made up of a separation means 21, a highorder data decoding means 22, a low-order data decoding means 23 and a decoding data generating means 24. Thus, the quality of the picture is made the same as that of the picture before compression.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and method for compressing and restoring image data such as natural images.

[0002]

2. Description of the Related Art Encoding / decoding of image data by a JPEG algorithm is known as one of methods for compressing the amount of image data. In the encoding process, this JPEG algorithm (baseline) divides image data into blocks of 8 × 8 pixels, and D blocks are divided into blocks.
After performing CT (discrete cosine transform), quantization and Huffman coding are performed. The decoding process in JPEG is the process of Huffman decoding, inverse quantization, and IDCT (discrete inverse cosine transform) according to the reverse procedure.

The above-mentioned quantization processing is performed based on a predetermined quantization table, and all the data within a predetermined range in the coefficient data after the DCT processing are converted into data of the same value. By thus reducing the number of digits of the coefficient data, the total data amount of the coefficient data is reduced.

[0004]

However, in such a conventional image data compression / decompression technique, since the above quantization processing is irreversible processing, the quantized data is converted into the data before quantization. It is impossible to restore. That is, the image quality of the restored image has deteriorated. Also,
There is a lot of complicated arithmetic processing such as DCT, which compresses image data
There was a problem that it took time to restore. In particular,
The image data for printing represented by Y, M, C, and BK has a huge amount of data, and is not necessarily suitable for data processing by the above method.

[0005]

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide an image data compression / decompression technique capable of shortening the time required for image data compression and decompression processing without degrading the decompressed image quality. Its purpose is.

[0006]

According to a first aspect of the present invention, an image is formed by arranging a plurality of pixels in a matrix and each pixel is compressed by a binary signal of a plurality of bits. In the device, an upper bit compressed data generating means for generating upper bit compressed data by performing run length compression on signals of upper bit parts of a plurality of pixels in the matrix of pixels for each equal bit sequence, and a lower bit of the pixel. By combining the signals of the side bit parts for a plurality of pixels in the matrix, lower bit compressed data generating means for generating lower bit compressed data, and compressed image data based on the upper bit compressed data and lower bit compressed data. An image data compression device comprising: a compressed image data generation unit.

According to a second aspect of the present invention, there is provided an image data decompression device for decompressing the compressed image data according to the first aspect, wherein the compressed image data is composed of upper bit compressed data and lower bit compressed data. Separation means for separating into
Upper side data decompression means for generating decompressed data of upper side bit parts of a plurality of pixels in the matrix of pixels based on the higher order bit compressed data, and a plurality of pixels in the matrix of pixels based on the lower order bit compressed data Lower-order data restoration means for generating restoration data for the lower-order bit portion of the, and restoration data for a plurality of pixels in a matrix based on the restoration data for the higher-order bit portion and the restoration data for the lower-order bit portion. And an image data restoring device including a restoration data generating unit.

According to a third aspect of the present invention, there is provided a method for compressing image data in which an image is formed by arranging a plurality of pixels in a matrix and each pixel is represented by a binary signal of a plurality of bits. A step of generating upper bit compressed data by performing run length compression on the upper bit portion signals of a plurality of pixels in the matrix of pixels for each equal bit string, and the signal of the lower bit portion of the pixel in the matrix In the method for compressing image data, the method comprises the steps of generating lower bit compressed data by grouping together a plurality of pixels and generating compressed image data based on the upper bit compressed data and the lower bit compressed data. is there.

A fourth aspect of the present invention is an image data restoration method for restoring the compressed image data according to the third aspect, wherein the compressed image data is converted into upper bit compressed data and lower bit compressed data. A step of separating, a step of generating decompressed data of high-order bit parts of a plurality of pixels in the matrix of pixels based on the high-order bit compressed data, and a step of generating decompressed data in the matrix of pixels based on the low-order bit compressed data. Generating restored data for the lower bit portion of the pixel, and generating restored data for a plurality of pixels in a matrix based on the restored data for the upper bit portion and the restored data for the lower bit portion. , Is a method for restoring image data.

[0010]

In the image data compressing apparatus according to the first aspect, a plurality of pixels are arranged in a matrix to compress the image data forming an image. Each pixel is represented by a multi-bit binary signal. The signals of the higher-order bit parts of a plurality of pixels in the matrix of pixels are run-length compressed for each equal bit string. As a result, high-order bit compressed data is generated. The signal of the lower bit part of the pixel is
A plurality of pixels are collected by the above matrix. As a result, lower bit compressed data is generated. Compressed image data is generated based on these high-order bit compressed data and low-order bit compressed data.

In the image data decompression apparatus according to the second aspect, in the plurality of pixels in the pixel matrix, the binary signal of the higher-order bit part is run-length compressed for each equal bit string, and the lower-order bit part of the two-bit signal is divided into two. The value signal restores the compressed image data that has been compressed by combining adjacent pixels. The compressed image data is separated into high-order bit compressed data and low-order bit compressed data, and each is individually restored. Then, the image data is restored based on the restored data of the higher-order bit portion and the restored data of the lower-order bit portion.

According to the image data compression method of the third aspect, the matrix binary image data is compressed. First, for a plurality of pixels, the upper bit part is run-length compressed for each equal bit. The remaining lower-order bit parts are grouped for a plurality of pixels. For example, when the upper 4 bits are run-length compressed, the lower 4 bits of one pixel and the lower 4 bits of pixels adjacent in the matrix in the column direction are treated as 1 byte.

According to the image data decompression method of the fourth aspect, the compressed image data generated by the above method is separated into the high-order bit compressed data and the low-order bit compressed data, and each is subjected to the compression method. Restore by the corresponding restoration method. Then, the restored image data is generated based on the upper restored data and the lower restored data.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS. In this embodiment, the present invention is applied to an apparatus for compressing and decompressing image data.

FIG. 1 is a block diagram showing a device for compressing image data. Reference numeral 11 is an upper bit compressed data creating means, 12 is a lower bit compressing means, and 13 is a compressed image data generating means. For example, in a 1-plane natural image formed by arranging a plurality of pixels in a matrix, adjacent pixels have a strong correlation (in most cases, substantially the same data continues). This image is divided into blocks of a matrix of a predetermined size, and image data is compressed and restored in units of blocks.
Further, the pixel can be represented by an 8-bit binary signal. Upper bit compressed data generation means 1
1 is to generate high-order bit compressed data by performing run-length compression on the signals of the high-order bit part of a plurality of pixels in the same column in the pixel matrix for each equal bit sequence. The lower bit compressed data generating means 12 generates lower bit compressed data by collecting the signals of the lower bit portion of the pixel for a plurality of adjacent pixels in the matrix. The compressed image data generation means 13
Compressed image data is generated based on the upper bit compressed data and the lower bit compressed data.

FIG. 2 is a block diagram showing the configuration of a decompression device for decompressing the compressed image data. That is, the decompression device separates the compressed image data into high-order bit compressed data and low-order bit compressed data, and based on the high-order bit compressed data, the high-order side of a plurality of pixels in the same column in the matrix of pixels. Higher-order data restoration means 22 for generating restoration data of bit part
And the lower side data decompression means 2 for generating decompressed data for the lower side bit parts of a plurality of pixels in the same column in the matrix of pixels based on the lower order bit compressed data.
3, and restored data generation means 24 for generating restored data of a plurality of pixels in a matrix based on the restored data of the upper bit portion and the restored data of the lower bit portion.
And are equipped with.

FIG. 3 shows a more specific structure of the image data compression apparatus. This image data compression device
It has an upper bit compression unit 31 for performing run length compression and a lower bit compression unit 32. Upper bit compression unit 31
Is capable of compressing the upper 4 bits of image data represented by an 8-bit binary signal. The high-order bit compression unit 31 uses the OR operation unit 311.
And an adding means 312 for adding the results thereof. For example, for the pixel (254) represented by "11111110", the logical sum (AND) with "10000000" is taken to examine the data of the 8th bit, and the result "10000000" and the same operation result "1" are obtained.
The number of adjacent pixels in the same column indicating "0000000" is added. That is, the run length of data in which the 8th bit is "1" is obtained.

The lower bit compression unit 32 includes a logical sum operation unit 321, a shift operation unit 322, and an addition unit 3.
23, and is comprised. OR operation means 32
1 is a pixel (2
54), the logical sum (AND) with “000011111” is taken to examine the data of the lower 4 bits. The shift operation means 322 shifts the operation result “00001110” to the left by 4 bits. This shift operation is performed only on odd-numbered pixels, and if the operation result is "11100000" and the next even-numbered OR operation result is "000011110", this operation result is added by the adding means 323 to obtain 8 Bit data “11101110” is created.

FIG. 4 shows the restoration (decompression) of the compressed image data.
2 shows an image data restoration device for performing the above. This device
An upper bit decompression unit 41 that decompresses upper 4-bit data from 8-bit compressed data, a lower bit decompression unit 42 that decompresses lower 4-bit data, and decompression by both of these decompression units 41, 42. Adding means 43 for adding the generated data to generate restored image data.

The high-order bit restoration section 41 has a shift operation means 411, an addition means 412, and a buffer 413. The shift calculation means 411 performs shift calculation on the compressed data. Adding means 412
Is to decompress the data compressed by the run length coding into the upper 4-bit data by performing addition. The decompressed high-order bit data is stored in the buffer 413 for 16 pixels, for example.

The lower bit restoration unit 42 has a logical sum operation means 421, a shift operation means 422, and a buffer 423.
And have. The logical sum operation means 421 is for performing decompression using a mask opposite to the mask used for the compression. That is, "11110" is added to the compressed data.
000 ". Then, the shift calculation means 422 reversely shifts only odd-numbered pixels,
A 4-bit right shift is performed. The result of this operation is
For example, 16 pixels are temporarily stored in the buffer 423. Restoration data of lower 4 bits of a pixel stored in the buffer 423, for example, "0000".
1110 ”is the upper 4-bit restored data“ 11110 ”for the same pixel stored in the buffer 413.
000 ”is added by the above-mentioned adding means 43, and the restored image data“ 11111110 ”is generated.

FIG. 11 is a flowchart showing the procedure for compressing image data in the above apparatus, and FIG. 12 is a flowchart showing the procedure for restoring compressed data. First, the procedure of compressing image data will be described.

The image data is compressed by first setting the number of pixels forming the image, the block size, the number of upper bits, and the number of lower bits (S1101). For example, the block size is 16 pixels, the number of upper bits is 4 bits, and the number of lower bits is 4 bits. Then read the image data (S
1102). For example, FIG. 5 shows a part of the read image data. In this figure, the data of each pixel is represented by the decimal system. In the pixel matrix, the second columns “254”, “254”, “252”, ...

Next, the upper bits are compressed (S110).
3). That is, the second column pixel is displayed in binary as shown in FIG. 6, and the upper 4 bits in the expanded state, for example, “254” is “11”.
11 ”(Bits 4 to 7) are subjected to run length compression for the same bit. In the case shown in the figure, when the block size is 16, that is, when 16 consecutive pixels are used as one unit, in Bits 7, 6 and 5, the upper bit compression code is "16, 0" with the number of 1s being 16 and the number of 0s being 0. Yes, Bit4
It is "9,7". Then, the high-order bit compressed data compressed in this way is output (S110).
4).

Next, the lower bits are compressed (S110).
5). For example, as shown in FIG. 7, the Bit of the pixel data
Regarding 0 to 3, the lower 4 bits of two adjacent pixels are grouped together in consecutive pixels. That is, if the lower 4 bits of the first pixel are “1110”, this and “1110” that is the second pixel are subjected to shift operation, addition, etc. to obtain one 8-bit data “11101”.
110 "(" 238 "in decimal notation). When arithmetic processing is performed on a plurality of pixels in this way, the lower bit compression codes “238”, “207”, “15”
1 ”... will be generated. Then, this lower bit compression code is output as lower bit compression data (S1106). As shown in FIG. 8, these compressed data are stored or transmitted in one block units. Then, this process is repeated until the compression is completed for all the image data (S1107).

Next, decompression and decompression of image data compressed by the above method will be described with reference to FIGS. First, the block size, the number of upper bits, the number of lower bits, etc. are set (S1201). Next, the compressed data is read (S1202). The compressed data is, for example, the data shown in FIG. Then, high-order bit decompression processing is performed on this compressed data (S1203). As shown in FIG. 9, high-order bit compression code (for 4 bits)
Is expanded to obtain high-order bit restored data of, for example, “1110000”.

Next, the lower bit restoration process is performed (S12).
04). For example, as shown in FIG. 10, the low-order bit compression code “238” is separated into “1110” + “1110” and “207” to “1100” + “1111” are generated. Then, the high-order bit restored data “11110000” and the low-order bit restored data “00001110” restored in this way are added (S).
1205). As a result, the restored image data “111111
10 ”is generated. Then, the restored image data is output (S1206). This process is repeated until all the pixels are completed (S1207). As described above, the restored image data is obtained from the compressed data. In this case, there is no data loss and the image quality is exactly equivalent to that of the original image before compression. Further, since complicated arithmetic processing such as in the case of DCT is not performed, the time required for the compression / decompression processing can be significantly shortened as compared with those.

In the case of a natural image, when each pixel of a blocked pixel array is decomposed into a bit array, the probability that the upper bits of adjacent pixels will change is low. Therefore, the continuous length of these bits is output as a high-order bit compression code. At the same time, the lower bit, that is, the number of bits capable of expressing one pixel (8 bits in the above embodiment), the number of bits obtained by subtracting the number of bits obtained by run length compression (4 in the above embodiment).
(Bit) is doubled and output as a low-order bit compression code of 8 bits. For example, if the upper 6 bits are compressed, the remaining 2 bits are combined into 4 pixels to generate the lower compressed code. As described above, in the present invention, the number of high-order bits for run length compression and the remaining number of low-order bits can be freely set and changed according to the type of the target image. Is. In the case of a color image, the compression and decompression of the image data is performed by the number of colors (R
This is performed for each color plane (3 colors for GB, 4 colors for YMCBk). Furthermore, the present invention is not limited to 8-bit data, but can be applied to 4-bit data, 16-bit data, and the like.

[0029]

According to the present invention, it is possible to shorten the time required for compression and decompression of image data. In addition, the image quality does not deteriorate due to compression and decompression. Further, according to the present invention, it is possible to reduce the storage capacity of the medium that stores image data.

[Brief description of drawings]

FIG. 1 is a block diagram showing an image data compression apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram showing an image data restoration device according to an embodiment of the present invention.

FIG. 3 is a block diagram showing an image data compression device according to an embodiment of the present invention.

FIG. 4 is a block diagram showing an image data restoration device according to an embodiment of the present invention.

FIG. 5 is a diagram showing data for explaining a procedure of compressing image data according to an embodiment of the present invention.

FIG. 6 is a diagram showing data for explaining a procedure of compressing image data according to an embodiment of the present invention.

FIG. 7 is a diagram showing data for explaining a procedure of compressing image data according to an embodiment of the present invention.

FIG. 8 is a diagram showing data when processing compressed image data according to an embodiment of the present invention.

FIG. 9 is a diagram showing data for explaining a procedure of restoring image data according to an embodiment of the present invention.

FIG. 10 is a diagram showing data for explaining a procedure of restoring image data according to an embodiment of the present invention.

FIG. 11 is a flowchart showing a compression process of image data according to an embodiment of the present invention.

FIG. 12 is a flowchart showing an image data restoration process according to an embodiment of the present invention.

[Explanation of symbols]

 11 Upper Bit Compressed Data Generation Means 12 Lower Bit Compressed Data Generation Means 13 Compressed Image Data Generation Means 21 Separation Means 22 Upper Side Data Restoration Means 23 Lower Side Data Restoration Means 24 Restoration Data Generation Means

Claims (4)

[Claims] 1. An image data compression apparatus in which an image is formed by arranging a plurality of pixels in a matrix, and each pixel is represented by a binary signal of a plurality of bits. Upper bit compressed data generating means for generating higher bit compressed data by performing run length compression on the signal of the side bit part for each equal bit string, and collecting the signal of the lower bit part of the pixel for a plurality of pixels in the matrix A lower bit compressed data generation unit that generates lower bit compressed data, and a compressed image data generation unit that generates compressed image data based on the higher bit compressed data and the lower bit compressed data. Image data compression device. 2. An image data decompression device for decompressing the compressed image data according to claim 1, wherein the decompression means separates the compressed image data into high-order bit compressed data and low-order bit compressed data. Upper side data decompressing means for generating decompressed data of upper side bit parts of a plurality of pixels in the matrix of pixels based on the higher order bit compressed data; and a plurality of pixels in the matrix of the pixels based on the lower order bit compressed data. Lower-order data restoration means for generating restoration data for the lower-order bit portion, and restoration data for a plurality of pixels in a matrix based on the restoration data for the higher-order bit portion and the restoration data for the lower-order bit portion. An image data restoration device comprising: restoration data generation means. 3. A method of compressing image data, wherein an image is formed by arranging a plurality of pixels in a matrix form, and each pixel is represented by a binary signal of a plurality of bits, wherein a plurality of pixels in the matrix of the pixels are provided. The step of generating high-order bit-compressed data by performing run length compression on the signal of the high-order bit part of each of the same-order bit strings, and combining the signals of the low-order bit part of the pixels for a plurality of pixels in the matrix, A method for compressing image data, comprising: a step of generating bit-compressed data; and a step of generating compressed image data based on the high-order bit compressed data and the low-order bit compressed data. 4. A method of decompressing compressed image data according to claim 3, wherein the compressed image data is separated into high-order bit compressed data and low-order bit compressed data. Generating decompressed data of upper bit parts of a plurality of pixels in the matrix of pixels based on the bit compressed data, and lower bit parts of a plurality of pixels in the matrix of pixels based on the lower bit compressed data And a step of generating restored data of a plurality of pixels in a matrix based on the restored data of the high-order side bit portion and the restored data of the low-order side bit portion. And how to restore the image data.