JPS6281880A - Data compression system - Google Patents

Abstract

PURPOSE:To maximize a compresson rate at every prescribed number of scanning lines and to reduce the transmission speed of a picture signal by allocating a code word, the higher the frequency of occurrence of an encoding mode included in the number of the prescribed scanning line, the shorter bit length of the code word. CONSTITUTION:A reading part 1 scans an image on an original and a detection part 2 detects the encoding mode with a two-dimensional encoding system from an obtained original image data and stores it at a buffer 7 at every scanning line in order. During the time, a frequency counting part 3 counts the frequency of occurrence of each encoding mode detected from the image data at every scanning line, then identifying its pattern, and selects a code chart corresponding to the pattern from a code chart storage part 4 with a code chart assigning part 5. And it is encoded according to a selected code chart at an encoder 6, and an encoded picture signal is transmitted to an opposite side.

Description

[Detailed Description of the Invention] [Summary] A data compression method used in facsimile machines that compress and transmit/receive image data using a two-dimensional encoding method, which encodes every predetermined number of lines, for example, one line of image data. This method aims to shorten the transmission speed by using short and long code words in the order of mote appearance frequency.

[Industrial application field]

The present invention relates to a data compression method provided in a facsimile machine.

A facsimile machine is a device that reads two-dimensional images pixel by pixel by scanning, converts them into electrical signals, and sends and receives them. In this case, the electrical signal obtained by scanning is a binary signal, and usually black pixels or white pixels, especially white pixels, are often continuous.

In this way, by replacing the length (run length) of consecutive pixels of the same type with a code, the amount of data to be sent and received can be significantly reduced. is compressed by encoding and sent and received.

Regarding encoding methods, there are one-dimensional compression methods that focus on a single scanning line, and two-dimensional compression methods that focus on the length of multiple scanning lines. Initially, each manufacturer had different encoding and communication methods. The procedures for '! It has also become possible to communicate with each other between facsimile machines made by different manufacturers.

At this time, it is desired that the communication speed between facsimile devices can be further reduced.

[Conventional technology]

Conventional two-dimensional encoding methods, such as the standard two-dimensional encoding method by CCITT, are shown in Table 1.

Nine types of encoding modes as shown in Table 1 are defined based on the two-dimensional relationship between pixels of original image data, and a code word such as "display" is always assigned to each encoding mode.

Table 1 Encoding mode Code word path CP ) 0001 Horizontal (H) 001 Vertical (V (01) 1 Vertical (VR (11) Oll Vertical (VR (21) 000011 Vertical (VR (3)
1) 0000011 vertical (VL(11) 0
10 Vertical (VL(2)) 000010 Vertical CVL(3
1) 0000010 [Problem to be Solved by the Invention] The conventional example described above has problems in each encoding mode when reading one page of an image in which the distribution of black pixels and white pixels is in a standard manner. The frequency of appearance is counted, and codewords with shorter pitch I-lengths are assigned to each encoding mode in descending order of frequency of appearance.

However, the appearance frequency of each scanning line is not necessarily in this order. For example, the encoding mode for a scan line with all white pixels is p, p, p, ... ■ (0) if there is a scan line with not all white pixels immediately before it, and the bus is Mode (P) is the most common.

For this reason, there is a problem in that high compression efficiency cannot necessarily be obtained for images such as horizontally written text.

That is, an object of the present invention is to improve data compression efficiency more than the conventional example and to shorten transmission time.

[Failure to solve the problem]

FIG. 1 is a diagram showing the principle of the present invention.

Reference numeral 1 denotes a reading unit that reads images on paper by scanning.

2 is from the original image data obtained by the reading unit 1,
A detection unit that detects an encoding mode using a two-dimensional encoding method.

Reference numeral 3 denotes a frequency counting unit that counts the appearance frequency of each predetermined number of scanning lines of the encoding mode detected by the detection unit 2 and identifies the types classified by the permutation of the appearance frequency of the encoding mode.

Reference numeral 4 denotes a code table storage unit that creates and stores in advance a code table for each type, in which code words with short bits a are associated with each encoding mode in descending order of appearance frequency.

5 is a code table selection unit that selects a code table corresponding to the type identified by the frequency counting unit 3 from the code table storage unit 4;

6 is a code table in which each encoding mode obtained in the detection unit 2 is recorded in the code table storage unit 4).

This is an encoder that performs encoding using the code table selected by the code table selection unit 5.

[Effect]

That is, depending on the frequency of occurrence of encoding modes included in a predetermined number of scan lines, for example, one scan line.

In the coding mode that appears frequently, coding is performed so that a code word with a short bit size is assigned, and the compression rate is maximized for each predetermined number of scanning lines, thereby reducing the transmission speed of the image signal.

If the appearance frequency of each encoding mode in one scanning line is, for example, 1
Assuming that the values are shown in (bl Ij) in Table 2.

In the conventional example, the code word in column (C) is used, so the bit length of the encoded image signal is 775 bits.

Table 2 (a) (bl (cl (d)
Encoding mode Appearance frequency Conventional example Invention path CP
) 10 0001 0000010
Horizontal (H) 15 001 000010 Vertical C
VfO)) 25 1 011 Vertical [V
L(t)) 30 010 1 vertical [vL
(2)]' 10 000010 00000
11 vertical (VL(31) 25 000001
0 010 Vertical (Vl)) 15 011
000011 Vertical (VI?(2)) 20
000011 0001 On the other hand, in the case of the present invention, since the code word in column (d) is used, the pitch I-fj of the encoded image signal becomes 655 bits, which can be reduced by more than 16%.

〔Example〕

FIG. 2 is a block diagram of an embodiment when applied to a facsimile machine, and includes the same reference numerals as the principle diagram of FIG. 1.

A buffer 7 temporarily stores the encoding mode of one scanning line obtained by the detection unit 2.

Reference numeral 8 denotes a control unit that controls communication control of 9 and other parts of 1.

9 is a modem device (MODEM).

10 is a network control circuit (NCU).

Reference numeral 11 denotes a buffer that temporarily stores the type of code table used to encode the image signal transmitted from the partner station.

Reference numeral 12 denotes a decoder that decodes the image signal transmitted from the partner station using a code table corresponding to the type stored in the buffer 11 among the code tables stored in the code table storage section 4.

A printer 13 outputs the decoded image data as an image.

Reference numeral 14 denotes a switching circuit that switches to either the code table specifying unit 5 or the buffer 11 depending on encoding during transmission or decoding during reception.

Further, 15 is a run-length code storage unit that stores run-length codes used in two-dimensional encoding.

At the time of transmission, the reading unit 1 scans the image on the document, the detection unit 2 detects the encoding mode based on the two-dimensional encoding method from the obtained original image data, and converts the detected encoding mode into 1. The data is stored in the buffer 7 in order for each scanning line.

During this time, the frequency counting section 3 counts the appearance frequency of each encoding mode detected from the image data for each scanning line and identifies the type, and the code table specifying section 5 A code table corresponding to the selected type is selected from the code table storage unit 4.

The encoder 6 encodes the encoding mode stored in the buffer 7 using the codebook selected by the codebook selection unit 5 from among the codebooks stored in the codebook storage unit 4, and converts the encoded image into a coded image. The signal is in the data format shown in Figure 3.
Each scanning line is assigned a type code and transmitted to the other party.

During reception, an image signal with a type code attached to each scanning line is similarly transmitted from the other station.

The decoder 12 decodes the encoded image signal using a corresponding code table, and the decoded image data is output by the printer 13.

As shown in Table 1, there are nine types of encoding modes and five types of code word bit lengths.

In other words, the total number of types of heels given for each scanning line is
The number of j1 columns in which 5 items are selected from each of 9 different items is 15,120 types, and is represented by a 14-bit code.

Therefore, according to the present invention, the compression ratio may deteriorate in some cases, and the code table storage unit 4 requires a large storage capacity. but.

vl in encoding mode, (3) ・VR(31・VL
(2) and V R (2+.

In particular, VL(3) and VR(3) appear very rarely.

By integrating the scanning line types including these encoding modes, the number of types can be reduced from 420 types to 20 types.

〔Effect of the invention〕

As explained above, according to the present invention, the data compression efficiency can be increased more than the conventional example, and the transmission time can be shortened.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the principle of the present invention. FIG. 2 is a configuration diagram of the embodiment. FIG. 3 is an explanatory diagram of the embodiment. In the figure. 1 is the reading section, 2 is the detection section. 3 is a frequency counting section, and 4 is a code table storage section.

Claims (1)

[Claims] A data compression method that compresses image data obtained by scanning using a two-dimensional encoding method and a code word, the method comprising: a frequency counting unit (3) that counts the appearance frequency for each predetermined number of scanning lines and identifies types categorized by the permutation of the appearance frequency of the coding modes; A data compression method characterized by comprising a code table storage unit (4) that creates and stores in advance a code table for each type in which code words with short bit lengths are associated with each other.