JPH0730764A - Method and device for compression - Google Patents

Abstract

PURPOSE:To improve the compressibility of image data. CONSTITUTION:When compressing the dot sequential image data of the respective eight bits of R, G and B, division is performed with three bytes as one block. The R data of respective blocks are collectively compressed, next, G data are collectively compressed and finally, B data are collectively compressed. Information on the number of blocks is defined as the header of compressed data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a compression method and apparatus, and more particularly to a compression method suitable for data having strong correlation periodically such as dot sequential RGB image data. And equipment.

[0002]

2. Description of the Related Art When compressing a series of data, it is usual to compress the data string in 1-byte units from the beginning. Decompression is the reverse process. This is usually
This is because the information is formed in units of 1 byte.

However, multi-valued image data, for example,
When the RGB image data is dot-sequential, three image data, R, G, and B, form an information unit. For example, R, G, and B are 8 bits each,
That is, when represented by 1 byte, the dot-sequential RGB image data represents 1 pixel by 3 bytes.

Image data has a high probability of having similar values between adjacent pixels, but in the case of the dot-sequential RGB format, if compression coding is performed in byte units, the expected compression ratio cannot be obtained. That is, next to the R data is G data that generally has a weak correlation with this R data value, and next to the G data is B data. This does not mean that the correlation between adjacent pixels is used.

The color representation format of image data is R
In addition to the GB format, there are luminance / color difference formats and the like, and the transmission (or storage) order of each component data includes the field order and the line order in addition to the dot order. A compression device that can realize a high compression rate corresponding to these various image formats is desired.

It is an object of the present invention to present a compression method and apparatus that meets such needs.

[0007]

A compression method according to the present invention divides a data string to be compressed into a plurality of blocks each having a predetermined number of elements and collectively compresses data at the same element position in the block. It is characterized by

The compressing apparatus according to the present invention comprises first means for dividing a data string to be compressed into a plurality of blocks each having a predetermined number of elements, and second means for collectively compressing data at the same element position of the block. And means of.

[0009]

By the above means, it is possible to perform compression by utilizing the regularity or periodicity of the data to be compressed. Therefore, a high compression rate can be achieved.

[0010]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows a schematic block diagram of an embodiment of the present invention. Reference numeral 10 denotes the arithmetic processing unit of this embodiment,
Compression means 12 consisting of hardware or software,
A decompression means 14 also made of hardware or software, a memory 1 for storing original data before compression.
6 and a memory 18 for storing compressed data, which are connected by an address / data bus 20 so that the compression means 12 and the decompression means 14 can access the memories 16, 18.

The original data before compression is transferred to and stored in the memory 16 from other application software or an image scanner, but may be directly supplied to the compression means 12. Is.

In the present invention, the data string of the data to be compressed is divided into units (blocks) suitable for sequential compression processing according to its characteristics, and the data is compressed as a block string. In each block, the data (element) of the basic unit belonging to each block is compressed using the elements located at the same position in the adjacent blocks. In other words, in the present invention, for a series of data strings, first, a cycle in which relevance can be expected is determined, and compression processing is performed between data of the same phase in that cycle.

In the present invention, the compression algorithm itself may be any, and in the present embodiment, the general Packbits method is used. Details of the Pack Bits method will be described later.

For example, if the original data is R, G, B
It is assumed that they are sequentially stored in the memory 16 in the 8-bit dot-sequential format, as shown in FIG. It should be noted that the data structure of the original data is known to the compression means 12 by the header information or the user's instruction.

Since each of R, G and B has a dot sequential format of 8 bits, one block has 24 bits, that is, 3 bytes. When the compression processing is in 1-byte units, the number of elements is 3.

First byte data in one block,
That is, R data is element # 1, second byte data, that is, G data is element # 2, third byte data.
Data, that is, B data will be referred to as element # 3.

Specifically, the compression process is performed on the same element of each block. That is, element # 1
Focusing on (that is, R data), all 01H (16
Since it is a decimal number, 01H will be repeated 5 times,
In the Pack Bits method, it can be represented as 04H01H as shown in the second byte and the third byte of FIG.

In element # 2 (that is, G data),
Since all are 02H (hexadecimal number), in the Packbits method, it can be represented as 04H02H, as shown in the 4th and 5th bytes of FIG.

Focusing on element # 3 (that is, B data), 03H is repeated three times, and then 04H and 05
Since H continues, it can be expressed as 02H03HFFH04H05H.

When decompressing the data thus compressed, information on the number of blocks of the original data is required. In this embodiment, the first byte of the compressed data is allocated for the block number information, and 05H is recorded as shown in FIG. 2 (2). In this embodiment, the amount of compressed data is 10 bytes. In addition to the information on the number of blocks, the information on the number of elements in one block is useful for decompression. Therefore, the element number information may be added to the compressed data string as a header or the like.

For reference, FIG. 2 (3) shows compressed data obtained by compressing the original data shown in FIG. 2 (1) by a conventional example. In the conventional example, the amount of data after compression is 1
It will be 6 bytes.

FIG. 3 shows a flow chart of the compression means 12 of this embodiment. First, the number of elements of the original data is determined, and the original data is divided into blocks (S1). The number of blocks is recorded at the beginning of the memory 18 (S2). The block number information is not limited to 1 byte and may be a predetermined number of bytes of 2 bytes or more.

A compression process is executed for the element # 1 of each block (S3), and the compressed data is stored in the memory 18 successively (S4). Element # 2 of each block
Is compressed (S5), the compressed data is continuously stored in the memory 18 (S6), the compression process is executed for the element # 3 of each block (S7), and the compressed data is stored in the memory. The data are sequentially stored in 18 (S8).

By the above processing, a compressed data string as shown in FIG. 2 (2) is generated.

FIG. 4 shows a flowchart of the expanding means 14. Substitute block number information into variable S (S21), 1
Control variables S1, S2, S for the number of block elements
3 is prepared and initialized with 0 (S22). Decompress the compressed data in order, store the number of decompressed data in the variable S1,
It is assumed that the decompressed data until 1 matches S is the data of element # 1 (S23, 24, 25). S1 is S
(S25), the number of data after decompression is stored in the variable S2, and the decompressed data until S2 matches S are the data of the element # 2 (S26, 27,
28). After S2 matches S (S28), the number of expanded data is stored in the variable S3, and the expanded data until S3 matches S is the data of element # 3 (S29, 30, 31). ). If S3 matches S (S3
0), the decompression process ends.

In this way, decompressed data can be obtained for each element. If the decompressed data for each element is stored in the memory 16 with a cycle of the number of elements of the block and with an offset of the number of symmetrical elements, decompressed data in the same sequence as the original data before compression can be obtained. it can.

The Packbits compression algorithm will be briefly described. When the length is 0 or more and less than 127,
The subsequent 1-byte data is continuous by the number obtained by adding 1 to the numerical value indicated by the length. Length is 1-12
When it is 7 or more and -1 or less, uncompressed data continues by the number obtained by adding 1 to the absolute value of the numerical value indicated by the length.

[0029]

As can be easily understood from the above description, according to the present invention, a high compression rate can be achieved when compressing data having periodicity.

[Brief description of drawings]

FIG. 1 is a schematic block diagram of an embodiment of the present invention.

FIG. 2 is an operation explanatory diagram of the present embodiment.

FIG. 3 is a flowchart of a compression unit 12.

FIG. 4 is a flowchart of the expanding means 14.

[Explanation of symbols]

10: arithmetic processing device 12: compression means 14: decompression means 16: original data memory 18: compressed data memory 20: address / data bus

Claims (2)

[Claims] 1. A compression method, wherein a data string to be compressed is divided into a plurality of blocks each having a predetermined number of elements, and data at the same element position in the block is collectively compressed. 2. A first means for dividing a data string to be compressed into a plurality of blocks each having a predetermined number of elements, and a second means for collectively compressing data at the same element position of the block. Compressor characterized by.