JPS59117864A - System for controlling data compression and expansion - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION Technical Field of the Invention The present invention relates to a modified read method which is an image data compression method. The present invention relates to a data compression/expansion control method in which an extension code is added after the EOL code to indicate the year and when uncompressed data is sent.

Prior Art and Problems When image data is transmitted, for example, by facsimile, there is a modified read (hereinafter referred to as MR) method as an international standard for compressing the image data. When transmitting image data using this MR method, in some cases uncompressed original data may be transmitted, but the amount of transmitted data may be small, and in such cases, uncompressed data, that is, original image data is transmitted. There is an uncompressed mode as an extended code that can be inserted directly into compressed data. This uncompressed mode is: ■ An identification word for entering uncompressed mode from an MR (MH) code ■ An expression of image data in uncompressed mode ■ An identification word for returning from uncompressed mode to a normal MR (MH) code It consists of three terms. Note that the MH code here is a modified Huffman code indicating the run length, and is 5MR.
It is used together with codes for data compression.

By the way, the following two identification words are used to enter the uncompressed mode from the MR (MH) code in (■) above, and * is "0".
” or “1” and in uncompressed mode, ***
=111.

Identification word in one-dimensional (MH) encoding...9. 000000001 ***Two-dimensional (MR)
Identification word in encoding... 0000001 *** However, in the case of MH code, this identification word has a terminating code (a code indicating the run length from 0 to 63) of 0.
It cannot be inserted after a code ending in 00. This is because it is confused with the line synchronization signal EOL (0000000001), so in such a case special processing such as delaying insertion by one line must be performed.

Further, the image data in the non-compressed mode in (2) above is shown in FIG. 1 (a). However, in order to distinguish it from the identification word for return in (2) described below, "1" is inserted when five consecutive rOJs occur so that six or more "0"s (white) do not continue.

The identification word that indicates when returning from the uncompressed mode to the MR code in (2) above is shown in FIG. 1 (b). The image data in FIG. 1 (b) is the last image data expressed in uncompressed mode, and T is a tag bit representing the next / color tone, and if the next run is black, T =
If rlj is white, T=“0”.

In order to determine in what cases data should be sent in uncompressed mode when image data is currently being sent out as compressed data, the conditions for switching to uncompressed mode are an issue. For example, from the simple method of switching when short runs of white and black alternate (for example, when there are white bits and black bits alternating like in a checkerboard pattern), to a certain long run When the number of bits when encoding the image data of S tL is ci, this difference di = ci - ti is calculated conditionally, and when this exceeds 20 bits of the identifier word, the mode is switched to uncompressed mode. is conventionally considered.

However, in the case of the uncompressed mode, 20 bits are required as an identification word, so it is not possible to frequently switch between the normal MR code and the uncompressed mode, and the key is to decide on a certain length li. In addition, conditions such as insertion of "1" into the image data in the uncompressed mode are added, making switching determination extremely difficult. When this is implemented using hardware, there is a drawback that the scale increases and calculation time is required to make the switching decision, making it unsuitable for high-speed processing.

Purpose of the Invention The purpose of the present invention is to improve the above-mentioned drawbacks by:
Compression or non-compression is determined for each line, and a tag bit indicating whether the next line is uncompressed data is added after the line synchronization signal EOL of the MR method and a tag bit indicating one-dimensional/two-dimensional distinction. Next, by continuing compressed data or uncompressed data (image data for one line), switching processing can be performed relatively easily, the compression ratio of one line will not be lower than 1, and it can be distinguished from EOL. An object of the present invention is to provide a data compression/expansion control method that does not require the insertion of "1".

Structure of the Invention To achieve this object, the data compression/expansion control method of the present invention includes a data compression section that compresses input image data, and a data compression section that compresses input image data. an input image data holding unit that holds input image data that does not exceed t″L; a data comparison unit that compares the amount of uncompressed data and compressed data for one line of image data; and a line synchronization signal and A code generator is provided for adding a tag bit for compressed information indicating whether the data of the next line is compressed data or uncompressed data after a tag bit indicating whether the next line is one-dimensional or two-dimensional encoded, The present invention is characterized in that a tag bit for compressed information is selectively added based on the determination result of the data amount comparing means, and then compressed data or uncompressed data of the eline is continued.

Embodiments of the Invention Before describing the present invention in detail in terms of one embodiment, an outline thereof will be explained.

In the present invention, in addition to holding uncompressed data for one life in a buffer, compressed data is created separately from this 15-inch data using the MR method. Determine whether the amount is larger or smaller, and if it is smaller, send compressed data.

Then, after the line synchronization signal EOL that is sent out prior to the data for the line, one-dimensional and two-dimensional tag pins 1lc4
Add one more compression/uncompression tag bit and send 0
Therefore, it is necessary to decode the next two tag bits of the EOL to identify whether the data to be transmitted next is compressed or uncompressed data and, in the case of compressed data, whether it is one-dimensional or two-dimensional compressed. Can be done.

That is, in the present invention, as shown in FIG.
A line synchronization signal and tag bit are added in the *, *, and J states. These symbols mean the following.

EOL code: 0000000001 *I: Tag bit indicating 1-dimensional or 2-dimensional,
rlJ for one dimension, "0" for two dimensions *2: Tag bit indicating compression or non-compression; "l" for compression, @"0" for non-compression. However, in the case of non-compression, 1 There is no distinction between dimension and two dimensions.

Therefore, the following can be recognized from the state of % *l p *.

OO: Uncompressed 01: Compressed in two dimensions 10: Uncompressed 11: Compressed in one dimension In this way, in the state of EOL+*,,*, as shown in Figure 2, it is determined whether the next data is compressed data or uncompressed data. Data can be identified.

An embodiment of the present invention will be described based on FIG.

In the figure, 1 is a data compression section, 2 is an uncompressed data node 7a, 3 is a compressed data buffer, 4 is a data comparison section, 5 is a code generation section, and 6 is a multiplexer.

The data compression unit 1 compresses the input image data for the line in one dimension or two dimensions, and compresses the input image data for the line in either mode according to the control signal from the control unit (not shown).
- It is used to create data.

The data amount comparison section 4 compares and determines whether the data amount of the compressed data for the era line created by the data compression section 1 is smaller or larger than the data amount of the uncompressed data. The uncompressed image data for the line is a fixed amount determined by the size of the document to be read by facsimile, for example, B4
In terms of size, the number of pixels per image is 2048 dots, and the amount of image data is constant at 2048 bits. Therefore, on the receiving side, even if a pseudo EOL exists in uncompressed data, it is image data and will be ignored.

In this way, the amount of uncompressed data is constant depending on the document size, and the constant L (2048 in the case of B4 size) according to the document size is sent from the control unit (not shown). The data amount comparator 4 can determine for each two inputs whether the data amount is larger or smaller than the uncompressed image data amount.

The code generator 5 generates the synchronization signal EOL and the following two tag signals F*I,*, and the data compression from the y'-time compression unit 1 is performed in one dimension, but it is performed in two dimensions. In addition, the data amount comparison unit 4 transmits determination information indicating which data amount is larger, compressed data or uncompressed data. Therefore, when compression is performed in one dimension, *1 is set to "1", and when compression is performed in two dimensions, *1 is set to rOJ, and when the amount of compressed data is small, data is transmitted in compressed mode. , *2 are set to "1", and otherwise *2 is set to "0" in order to send uncompressed data.

The multiplexer 6 outputs the image data in the state shown in FIG.
+, *t is output. Next, when a signal indicating that the amount of compressed data is small is transmitted from the data amount comparator 4, the compressed data transmitted from the compressed data buffer 3 is output, and in the opposite case, the compressed data transmitted from the uncompressed data buffer 2 is output. Output uncompressed data.

Next, the operation shown in FIG. 3 will be briefly explained.

When image data for one line of a document is input from the document reading section (not shown), this is transferred to the uncompressed data buffer 2.
At the same time, compressed data is created by the data compression unit 1. And this compressed data is compressed data/
The data is sent to the Jtsufa 3 and held. At this time, the amount of 1'-in compressed data output to the 4-stock compressed data buffer 3 of the data company comparison section is counted. Accordingly, the data amount comparator 4 determines whether the amount of compressed data is smaller than the amount of uncompressed data. When it is determined that there is less, the determination result is sent to the code generator 5 and the multiplexer 6, and following goL+r*, *, J, the multiplexer 6 outputs the compressed data held in the compressed data buffer 3. become. Conversely, when the amount of compressed data is small, the multiplexer 6 outputs the image data held in the uncompressed data buffer 2 following EOL +r*+ *2 J transmitted from the code generator 5.

Effects of the Invention By using the extended code according to the present invention, it is possible to determine whether it is compressed or uncompressed in one line, so unlike the conventional uncompressed mode, it is not necessary to use an identifier or insert a "1", so it is easier to identify and The results can be easily controlled. Moreover, in order to create the extended code of the present invention, it is only necessary to add an uncompressed data buffer, a compressed data buffer, a data amount comparison section, etc. to a conventional data compression circuit, and it is extremely easy to determine whether it is compressed or not. be. In addition, since each of the uncompressed data buffer and compressed data buffer has a memory capacity for one line, it can operate at the same processing speed as a conventional data compression circuit that does not have an uncompressed mode, making it suitable for high-speed processing. .

[Brief explanation of the drawing]

FIG. 1(a) shows the representation of image data in uncompressed mode, FIG. 1(b) shows the identification word for returning to MR code from uncompressed mode, and FIG. As an example, FIG. 3 shows the configuration of an embodiment of the present invention. In the figure, 1 is a data compression unit, 2 is an uncompressed data buffer,
3 is a compressed data buffer, 4 is a data amount comparator, 5 is a code generator, and 6 is a multiplexer. Patent applicant Fujitsu Limited

Claims (1)

[Claims] In a data compression control method that compresses image data using the modified read method, there is a data compression section that compresses input image data, an input image data holding section that holds uncompressed input image data, and one line of image data. data amount comparing means for comparing the amount of data between uncompressed data and compressed data; A code generator is provided to add a tag bit for compressed information indicating whether the data is compressed data or uncompressed data, and the tag bit for compressed information is selectively added based on the determination result of the data amount comparing means. 1. A data compression/expansion control method, characterized in that compressed data or uncompressed data is continued.