JPS6338152B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (1) Technical Field of the Invention The present invention relates to a method for compressing and encoding an image signal using a modified Huffman (MH) method or a modified read (MR) method.

(2) Prior Art Conventionally, the Modified Huffman (MH) method or the Modified Read (MR) method specified in CCITT Recommendation T.4 has been frequently used as a compression method for facsimile encoding. FIG. 1 shows one example of this. After storing the serial image signal from the reading circuit 1 in the one-line divided signal buffer memory 2, in the MH method, the encoded line is read out from the image signal buffer memory 2, and the change point detection circuit 3 In white,
A black change point is detected and the microprocessor (MPU) 4 is notified of it. During this time, a run length number is created by the control circuit 5 such as a horizontal counter, given as an address to the code ROM table 6, and the output is sent to the P/S conversion circuit 7 and the MPU.
4, the MH code is converted into an effective bit length MH code, and after parallel/serial conversion is performed, the MH code is sequentially written into the code buffer memory 8.

Also, like the MH method, a compressed image signal is created using the modified read (MR) method. In the MR method, image signal buffer memory 2
The image signals of the reference line and the encoded line are simultaneously read out from the MPU 4, the change point detection circuit 3 detects the change point, and notifies the MPU 4 of the change point. During this time, the control circuit 5 such as horizontal and vertical counters detects the state of the change point, creates each MR encoding mode, gives it as the address of the encoding ROM table 6, and sends the output to the P/S conversion circuit. 7 and MPU
4, the code is converted into an MR code with an effective bit length, subjected to parallel/serial conversion, and then sequentially written into the code buffer memory 8. Furthermore, line synchronization signal, fill signal MPU4
encoded ROM table 6 through control circuit 5
and similarly written to the code buffer memory 8 after P/S conversion.

In this way, the conventional method stores and reads one line of serial image signals regardless of the amount of information in the image signal, and further writes encoded data serially to the code buffer memory, resulting in a slow processing speed. Therefore, it is unsuitable for high-speed processing. In addition, in the conventional system, the means for encoding each of the MH and MR codes is constituted by hardware and is complicated, and these encoding systems have a fixed hardware configuration and are not versatile.

(3) Purpose of the Invention The present invention eliminates the above-mentioned drawbacks, and uses a method of compressing and encoding image signals using the MH method or MR method, and provides versatile and high-speed encoding with a simple hardware configuration. The purpose of the present invention is to provide an image signal encoding method suitable for

(4) Structure of the Invention In order to achieve the above object, the image signal encoding method of the present invention has an encoding table based on the modified Huffman (MH) method or the modified read (MR) method and compresses the image signal. In the coding method, the method includes means for detecting the position of a change point in one line of image signal and storing the position of the change point as an absolute address, and performing arithmetic processing to determine the corresponding position in the encoding table of the MH method or MR method. It is characterized in that encoded data is obtained by converting it into a code.

(5) Embodiment of the invention FIG. 2 is an explanatory diagram of the configuration of an embodiment of the invention.
In this figure, components having the same functions as those in FIG. 1 are labeled with the same numbers even if the order of insertion is different from that in FIG. 1.

In the figure, the serial image signal from the reading circuit 1 is directly input to the change point detection circuit 3 to detect the change points of white and black, and the corresponding change point position counter 10
The count value is stored in the image signal buffer memory 2 as an absolute address. The absolute address is a continuous value from 1 to 2048, for example, and indicates the position on one line at which the change point of the image signal is located.

Further, the movement of the address of the image signal buffer memory 2 indicates the existence of a change point and the change between white and black of the image signal.

FIGS. 3a and 3b are operational waveform diagrams of the main parts of the embodiment. When the clock (CLK) applied to the change point detection circuit 3 and the change point position counter 10 for one line shown in figure a is shown in figure a, the change point of the read image signal in figure a is determined as the change point in figure a. A change point pulse is detected, and in response to this, the count value from the change point position counter 10, that is, the absolute address, is sent to the image signal buffer memory 2. FIG. 2b shows an example of the encoded lines stored in the image signal buffer memory 2 in this case. When using the MH method, the encoded line stored in the image signal buffer memory 2 is read out, and the MPU 4 performs arithmetic processing on the absolute address of the encoded line to create white (W) and black (B) as shown in FIG. Convert to the number of run lengths of MH
encoding in which the code and bit length of
Output as the address of ROM table 6. In facsimile encoding, the beginning always starts with a white run length, and the next change point is a black run length. Therefore, the output from the encoded ROM table 6, which includes redundancy, is converted into an effective bit length by the MPU 4.
It is converted into MH code and encoded, and the parallel outputs are sequentially packed and written into the code buffer memory 8. After performing the above operations for one encoded line, the process moves on to encoding the next encoded line in sequence.

In addition, the line synchronization signal and fill signal are MPU4
It is given from the encoded ROM table 6 according to instructions from the MH code, and is handled in the same way as the MH code. The output of the code buffer memory 8 is subjected to parallel/serial conversion by the P/S conversion circuit 7, and code data is output.

When using the MR method, the 5th point where the change point is stored in absolute address format from the image signal buffer memory 2
The previous reference line in Figure a and the current encoded line in Figure b are read out, the MPU 4 performs arithmetic processing on the encoded line based on the reference line, and provides it as an address in the encoding ROM table 6. We obtain MR codes as shown in some examples.

The MR code is expressed as a relative sign of the encoded line with respect to the reference line: V ₍₀₎ [move directly below], P [skip], V _L(1) [shift 1 bit to the left], V _R(1) [ Shift 1 bit to the right], H [shift 4 bits or more]...
etc.

(6) Effects of the Invention As explained above, according to the present invention, only necessary change point information is stored in absolute address format, so it is not necessary to store and read out one line's worth of serial image signals. Furthermore, since the encoded data is sequentially output in parallel and stored in the code buffer memory, high-speed encoding becomes possible. In addition, since MH and MR encoding is mainly based on a software configuration that creates addresses by detecting change points and calculating the number of white or black run lengths, a fixed hardware configuration is not required, and the entire hardware configuration is not required. Become. Therefore,
With the same hardware configuration, by changing the software processing, it is possible to perform not only other encoding but also decoding, and even access the image signal buffer memory from the MPU, so it is expected to expand into applications other than encoding and decoding.

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram of the configuration of the conventional example, Fig. 2 is an explanatory diagram of the configuration of the embodiment of the present invention, Figs. 3a and b are explanatory diagrams of the operation of the main parts of the embodiment, Figs. , b are other operation explanatory diagrams of the main parts of the embodiment, in which 1 is a reading circuit, 2 is an image signal buffer memory, 3 is a change point detection circuit, 4 is a microprocessor (MPU),
Reference numeral 6 indicates an encoding ROM table, 7 a P/S conversion circuit, 8 a code buffer memory, and 10 a change point position counter.

Claims (1)

[Claims] 1. In a method that compresses and encodes an image signal by having an encoding table based on the Modified Huffman (MH) method or the Modified Read (MR) method, detecting the position of a change point in the image signal of one line, It is equipped with a means for storing the change point position as an absolute address, and the above-mentioned MH method or
An image signal encoding method characterized in that encoded data is obtained by converting to a corresponding code in an MR method encoding table.