JPS6367392B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a system for multiplexing redundancy compression encoding of facsimile signals of a plurality of lines.

Modified Huffman encoding method (MH encoding method) is used as a redundancy compression encoding method for facsimile signals (see "Basics and Applications of Facsimile New Edition" edited by Institute of Electronics and Communication Engineers (published September 10, 1981)) or modified read. An encoding method (MR encoding method) is known. These redundancy compression encoding methods utilize the one-dimensional or two-dimensional statistical properties of facsimile signals to remove redundancy. For example, in the MH encoding method, continuous white or black on one scanning line is Redundancy is removed by Huffman encoding the length (run) based on the respective statistical properties of white or black. As is well-known, Huffman encoding is an encoding method that assigns variable length codes according to the frequency of appearance of a certain state, and the MH encoding method based on this encoding principle also assigns variable length codes depending on the length of white or black runs. A variable length code is assigned.

Therefore, when an original image signal of one scanning line is MH encoded, the code length varies depending on the pattern of the scanning line.

No problems occur when an encoder and decoder that performs such redundancy compression encoding operate on original signals from a single line, but when simultaneously encoding and decoding facsimile signals from multiple lines, is problematic. In other words, when processing multiple lines, there are cases in which input signals or output signals are processed bit by bit and cases in which they are processed in scanning line units, but in either case, the code length corresponding to one scanning line is the same. Since there is no encoder, the encoder operation is interrupted midway and processing of another line is started.
In this case, in bit-by-bit processing, the encoder's internal state is stored for the number of lines, and each time the own line is specified, the internal state is restored and processing continues, which makes the encoder complicated. At the same time, it is necessary to perform line switching control on a bit-by-bit basis. On the other hand, in the case of scanning line by scanning line, it is necessary to store the internal state, but line switching control may be carried out by scanning line.

However, it is complicated to retrieve or store all the internal states of the encoder;
It's not really a good idea.

SUMMARY OF THE INVENTION An object of the present invention is to provide a multiplex facsimile signal encoding system which eliminates the above-mentioned drawbacks and allows multiple lines to be processed multiplexed without storing internal states.

The method of the present invention is a multiplex facsimile signal encoding method in which redundancy compression encoding of facsimile signals of a plurality of lines is multiplexed. When the number of bits of each redundancy compression code corresponding to the code is not equal to an integer multiple of a predetermined number, a predetermined bit pattern is added to the code to make the code an integer multiple of the predetermined number N (positive integer). means and
original image signal storage means for storing the original image signals of the plurality of scanning lines of each of the lines; a storage area is divided into a plurality of areas corresponding to the number of lines, and each of the divided areas is divided into a plurality of blocks. a compression code storage means for storing a redundancy compression code given from the encoding means; and an addressing means for specifying an address in the compression code storage means in units of N bits; The original image signals of each of the scanning lines corresponding to the image data are respectively encoded into the compressed code by the encoding means, and each compressed code is sequentially stored in one block in the compressed code storage means, and the next block and When at least one compressed code is stored until a boundary is reached or the boundary is crossed, processing of this line is interrupted and an address is stored in the addressing means so that the next block can be stored continuously when processing is resumed. After that, process the next line.

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

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

The original image signal is the original image buffer 1 installed for line support.
is stored as shown in FIG. 2a. Similarly, the redundancy compression encoded code is stored in the code buffer 3 as shown in FIG. 2b. That is, as shown in FIG. 2a, the original image signal for one scanning line stored in the area of buffer 1 is redundancy compression encoded, and as shown in FIG.
The data is stored in the area of the first block within. Similarly, the original image signals of each scanning line stored in each area ~ in the buffer 1 are respectively encoded and stored in the areas ~ in the buffer 3. Here, the original picture buffer 1 and the code buffer 3 are logically divided into n blocks and m blocks, respectively. In this embodiment, each block of the original picture buffer 1 has a storage capacity for four scanning lines, with one scanning line's worth of original picture signals as a unit. Each block of the code buffer contains the maximum code length when the original image signal for one scanning line is encoded with redundancy compression (for example, in the case of MH encoding, the original image in which 1 white dot and 1 black dot are generated alternately) When it is a signal, the maximum code length is 974 bytes, assuming the number of dots in one scanning line is 1728)
It suffices to have a storage capacity of more than that.

Now, if an original image signal exists in the original image buffer 1 and two or more blocks of the code buffer 3 are empty areas, each original image signal is transferred to the encoder 2 according to an instruction from the control section 4.
The data is redundancy-compressed encoded and written to the code buffer 3. FIG. 3 shows the internal structure of the encoder 2, with terminals 20, 22 and 27 connected to the original picture buffer 1, the control section 4 and the code buffer 3, respectively. The original image signal inputted from the terminal 20 is given to a change point detection circuit 21 and a counting circuit 24. The counting circuit 24 performs counting until the change point detection circuit 21 finds a white or black change point in the original signal, and provides the count value when the change point is found, that is, the run length, to the MH encoder 25. The MH encoder 25 performs MH encoding on this run length. The compressed code from the MH encoder 25 is sent to the fill bit insertion circuit 26.
It is output to terminal 27 via. The fill bit insertion circuit 26 counts the code length of each compressed code from the encoder 25, and continues counting until the control unit 23 detects a scanning line break. When this delimiter is detected, the insertion circuit 26 checks the count value, that is, the total code length output so far, and inserts a fill bit (for example, logical “0”) to terminal 27
and initializes the count value for the next scan line. That is, all encoded compression codes of each scanning line are written into the buffer 3 with a code length that is an integral multiple of the integer N.

In this way, the compressed code is written into the buffer 3 until the compressed code reaches or crosses the boundary with the next block. That is, the second
In Figures a and b, the encoding process for this line is performed until the compression code of the original image signal of the scanning line stored in the area of buffer 1 is written in the area of buffer 3, and then the code of the next line is Move on to conversion processing. When transferring to this next line,
The address next to the last address of the area, that is, the first address of the area, is stored in the control unit 4. This address specification can be performed in units of N bits since the length of each compression code is an integral multiple of N. When the processing of another line is finished and the encoding process of this line is continued again, writing is performed from the first address of the area. Also, when the (m-1)th block is written, the contents of buffer 3 of this line (the contents of the 1st to (m-1)th blocks and the amount written beyond the m-th block) are transferred to the terminal 5. It is output to a facsimile terminal etc. via . In this way, the encoding process for each line is performed continuously and in a time-division manner.

Note that decoding may be performed by performing the opposite operation to that described above.

As described above, the present invention has the advantage that it is possible to easily perform multiple processing of encoding for a plurality of lines.

[Brief explanation of drawings]

FIG. 1 is a block diagram for explaining one embodiment of the present invention, FIGS. 2a and 2b are diagrams for explaining the storage formats of the original picture buffer and code buffer, respectively, and FIG. 3 is the configuration of the encoder 2. FIG. In the figure, 1...original picture buffer, 2...encoder, 3...code buffer, 4...control unit, 20...
...Input terminal, 21...Changing point detection circuit, 22...
Input terminal, 23... control circuit, 24... counting circuit, 25... MH encoder, 26... file insertion circuit, 27... output terminal.

Claims (1)

[Claims] 1. In a multiplex facsimile signal encoding method that multiplexes redundancy compression encoding of facsimile signals of a plurality of lines, each original signal corresponding to each scanning line is redundancy compression encoded, and each of the above-mentioned signals corresponding to each scanning line is encoding means that adds a predetermined bit pattern when each number of bits of the redundancy compression code is not equal to an integer multiple of the predetermined number N (positive integer); an original image signal storage means for storing original image signals of the plurality of scanning lines of each line; a storage area is divided into a plurality of areas corresponding to the number of lines; each area divided into the plurality of areas is divided into a plurality of blocks, respectively; one of the plurality of lines, comprising a compression code storage means for storing a redundancy compression code given from the encoding means, and an addressing means for specifying an address in the compression code storage means in units of N bits; The original image signal of each scanning line corresponding to the original image signal is encoded into the compressed code by the encoding means, each compressed code is sequentially stored in one block in the compressed code storage means, and the boundary with the next block is When at least one compressed code is stored until reaching or exceeding this boundary, the processing of this line is interrupted, and an address is stored in the addressing means so that the next block can be continuously stored when processing is resumed. A multiplex facsimile signal encoding method that is characterized by processing the next line.