JPH0332279A - Encoding system for facsimile equipment - Google Patents

Abstract

PURPOSE:To improve efficiency for encoding and to accelerate the speed of communication by executing two-dimensional encoding only concerning the first change picture element position of each encoding line. CONSTITUTION:A detecting means 111 detects the change picture element position of each line and inputs the first change picture element position of each line to an encoding deciding means 131. The encoding deciding means 131 compares the first change picture element position, which is stored in a storing means 121, of the referring line with this inputted first change picture element position of the encoding line. For example, when difference between these respective change picture element positions is within a prescribed value, it is judged that the two-dimensional encoding can be executed. In such a case, an encoding means 141 executes the two-dimensional encoding to the first change picture element position, which is detected by the detecting means 111, of the encoding line. Thus, the encoding can be efficiently executed.

Description

[Detailed description of the invention] 〔table of contents〕 overview Industrial applications Conventional technology Problems that the invention aims to solve Means to solve problems action Example ■.

■.

Correspondence between the embodiment and FIG. 1 Structure and operation of the embodiment (i) Overall structure of the facsimile machine (ii) Encoding process (iii) Decoding process ■ Summary of the embodiment ■1 Modifications of the invention Effects of the Invention [1 Already Required] Regarding the encoding method of the G3 facsimile machine, which efficiently encodes the white pixel at the beginning of each line, it is possible to increase the communication speed and reduce the cost of the device. a detection means for detecting a changed pixel position for each encoded line of a transmitted document; a storage means for storing the first changed pixel position of a reference line; and a first change in each encoded line detected by the detection means. An encoding determination means compares the pixel position with the value stored in the storage means to determine whether two-dimensional encoding is possible, and the changed pixel position detected by the detection means is input, and encoding means that performs two-dimensional encoding when two-dimensional encoding is possible for the changed pixel position of this encoding line, and performs one-dimensional encoding for other changed pixel positions of this encoding line. , so that two-dimensional encoding is performed only on the first changed pixel position of each encoding line.

[Industrial application field]

The present invention relates to an encoding system for a G3 facsimile machine that efficiently encodes white pixels at the beginning of each line.

In recent years, G3 facsimile machine covers 1' have become popular, and there is a demand for lower prices. Therefore, by reducing the functions that were conventionally achieved by the Pardo machine as much as possible, and by realizing the same functions by software processing on the processor that controls the facsimile machine, we have created a simple hardware system. Lower prices are expected.

On the other hand, even such low-cost fudging equipment is required to have higher performance, and in particular, in order to reduce line usage costs, it is necessary to improve the compression efficiency of image information and shorten communication time. be.

[Conventional technology]

The encoding methods of G3 facsimile machines generally include one-dimensional encoding methods and two-dimensional encoding methods standardized by CCITT Recommendations T and 4.

In the 1±L Nikai Gokasa decoy hole one-dimensional encoding method, one line of data is constructed by alternately combining a white run length code representing the number of white pixels and a black run length code representing the number of black pixels. All data lines begin with the sign of a white run, or if the actual line begins with a black run, then with a white run of length O.

There are also two types of codes: tarnishing codewords and makeup codewords. Run lengths from 0 to 63 pixels are encoded with only the appropriate or ending codeword. On the other hand, run lengths of 64 pixels or more are first encoded with a makeup code word representing a run length that is equal to or shorter than the run length.

This is followed by a terminating codeword that represents the difference between the actual runlength and the runlength represented by the makeup codeword.

Furthermore, before the first data line and after the data of each line, there is a line termination code EOL (Cheetah for my 11;!llo O0000000) indicating the end of the line.
001'") is added. Also, a fill FILL can be inserted between one line of data and the line end code EOL. These F, IL, and L are
・This is redundant data that is a variable length signal string of “0”°, and is not subject to decoding, but is used for time adjustment etc. according to the standard.

The end of one page is six consecutive line termination codes EOI
4' The configured control is indicated by the IH return code RTC.

In the two-dimensional encoding method, the position of each changing pixel on the current line, that is, the encoding line, is on the encoding line or on the reference line located immediately before the encoding line). This is a sequential coding method in which coding is performed in accordance with the position of a reference pixel. After a coding line is coded, the coding line becomes a reference Ilq line for the next coding line.

A changed pixel is defined as a pixel whose color, ie, white or black, is different from the color of the immediately preceding pixel on the same scanning line. ao l
al + aZ + bl + bZ is defined as follows.

ao: Reference or origin change pixel on the encoding line al: First change pixel to the right of ao on the encoding line a2; First change pixel to the right of al on the encoding line bI: To the right of ao and with ao The first change pixel on the reference line with the opposite color b2: The first change pixel to the right of the reference line There are encoding modes.

■Pass mode: When b2 exists on the left side of al, it is encoded in bus mode. When encoding in this mode, ao is set to a pixel on the encoding line immediately below b2 in preparation for the next encoding.

(2) Vertical mode If the relative position between al and al is 3 pixels or less, encoding is performed in vertical mode. After vertical mode encoding, the position of ao is moved to al.

■Horizontal mode: In this mode, the run length is a. a, and ao2 as the code word H+M (ao at ) +M (ao2
). Here, H is CCITT recommendation T
, 4 is a horizontal mode code based on a two-dimensional code table (not shown). M (ao al) and M (alaz)
are one-dimensional codes indicating the length and color of runs ao a and at az, respectively. After horizontal mote encoding, the position of ao is moved to a2.

End-of-line code EO at the end of every encoded line
L is added. After E OLO, the next line contains additional bits indicating whether one-dimensional or two-dimensional encoding is used. For example, if the next line is a one-dimensional encoding method, use E OL + 1, and if the next line is a two-dimensional encoding method, use E OL + 1.
Use OL+0. Note that EOL+1 is used before the first line of the page.

Similar to the one-dimensional encoding method, l-line data and EOL
A fill FILL can be inserted between. Also, the end of one page is six consecutive (EOL+1)
It is indicated by a control return code RTC consisting of:

[Problem to be solved by the invention]

By the way, when one-dimensional encoding is performed by software processing, there is a run-length counter, that is, a process that detects the color and length of a pixel (run length), and a one-dimensional code table for replacing the detected run length with a code corresponding to the detected run length. This can be easily realized by code conversion processing. Therefore, although it can be implemented on an inexpensive one-chip microcomputer, the compression efficiency is inferior to that of the two-dimensional encoding method, so there is a problem that it is not possible to increase the communication speed even remotely.

On the other hand, when two-dimensional encoding is realized only by software processing, it is necessary to calculate the changed pixel position and relative distance of both the reference line and the encoded line for each changed pixel point, which increases the complexity of the process and increases the processing time. It is not practical because it costs a lot of money. Therefore, in order to speed up the processing, hardware assistance is required, and there is a problem that it is not possible to reduce the price of the device.

The present invention has been created in view of the above points, and an object of the present invention is to provide an encoding method for a facsimile machine that can realize higher communication speed and lower cost of the device.

[Means to solve the problem]

FIG. 1 is a basic block diagram of the encoding method of a facsimile apparatus according to the present invention.

In the figure, a detection means 111 detects the position of a changed pixel for each encoded line of the transmission document.

The storage means 121 stores the first change image 0 of the reference line. Stores the raw position.

The encoding determination means 131 stores the first changed pixel position of each encoded line detected by the detection means Ill and the storage means 121.
It is determined whether two-dimensional encoding is possible by comparing with the stored value.

The encoding means 141 receives the changed pixel position detected by the detecting means 111, performs two-dimensional encoding on the first changed pixel position of each encoding line if it is possible, and performs two-dimensional encoding on the first changed pixel position of each encoding line. One-dimensional encoding is performed for other changed pixel positions in the encoding line.

Therefore, the overall configuration is such that two-dimensional encoding is performed only on the first changed pixel position of each encoding line.

[For production]

The detection means 111 detects the changed pixel position of each line,
Encoding determination means 131 for determining the first changed pixel position of each line
Enter. In the encoding determination means 131, the storage means 12
Compare the first changed pixel position of the reference line stored in 1 with the first changed pixel position of this input encoded line, for example, if the difference between these respective changed pixel positions is within a predetermined value set. It is determined that two-dimensional encoding is possible.

In this case, the encoding means 141 performs two-dimensional encoding by marking the position of the first changed pixel of the encoding line detected by the detection means 111. Furthermore, the encoding means 141 performs one-dimensional encoding when two-dimensional encoding is impossible or for changed pixel positions other than the beginning of the encoding line.

In the present invention, efficient encoding can be performed by performing two-dimensional encoding on the first pixel change position of each encoding line. In general, original documents sent and received between facsimile machines almost always have characters or figures drawn on a white background, and include continuous white areas at the left edge, top edge, bottom edge, or between lines. Therefore, by two-dimensionally encoding these, it is possible to increase the compression rate.

Further, since only the first changed pixel position of each of the reference line and the encoded line is compared to determine whether or not two-dimensional encoding is possible, implementation by software processing is also possible.

〔Example〕

Hereinafter, embodiments of the present invention will be described in detail based on the drawings.

FIG. 2 shows the configuration of an embodiment to which the facsimile apparatus encoding method of the present invention is applied. Further, FIG. 3 shows the configuration of an encoding section of an embodiment.

(1) Here, the correspondence between the embodiment of the present invention and FIG. 1 will be shown.

The detection means 111 corresponds to the run length counter 313.

The storage means 121 corresponds to the -time save register 315.

The encoding determining means 131 corresponds to the vertical mode determining section 317.

The encoding means 141 includes a code conversion section 319. This corresponds to code tables 321 and 323.

Examples of the present invention will be described below assuming that the correspondence relationship as described above exists.

In FIG. 2, 210 indicates a facsimile machine, 211
213 is the CPU (central processing unit), 213 is the reading section, and 2 is the reading unit.
15 is a recording section, 217 is a memory section, 219 is an operation panel section, 221 is an encoding section, 231 is a decoding section,
241 represents a modem, and 243 represents a line control unit.

The CPU 211 controls the entire facsimile machine 210, and performs control operations by executing a program stored in a memory section 217 composed of RAM or ROM. Furthermore, the user of facsimile device 210 can input instructions to CPU 211 by operating each operation key on operation panel 219 .

The reading unit 213 reads the image information of the transmitted original, and the recording unit 215 prints the received original on a predetermined recording paper.

The encoding unit 221 encodes and compresses the image information of the transmission document read by the reading unit 213, and the compressed data is sent to the modem 241.

The decoding unit 231 restores the image information (compressed data) of the received original, and the restored image information is sent to the recording unit 215 and printed on recording paper.

Modem 241 modulates and demodulates data.

The modulated transmission data is sent to an external line under the control of the line control section 243. Further, data input via an external line is received by the control section 9 of the line control section 243, and then demodulated by the modem 241 and sent to the decoding section 231.

Each component except the line control section 243 is connected to a bus, and the modem 241 is connected to an external line via the line control section 243.

Further, the facsimile device 210 can communicate with other facsimile devices 210 via a public line such as a telephone line or a private line.
0, and documents to be facsimile are sent and received. Note that the configuration of facsimile device 290 is assumed to be similar to that of facsimile device 210, and detailed description of the configuration will be omitted.

” 11 simple explanations of menstruation FIG. 3 shows a detailed configuration of the encoding section 221 described above.

In Figure 3, 311.325 is the main part, 313
315 is a temporary save register, 317 is a vertical mode determination section, 319 is a code conversion section, and 321 and 323 are code tables.

The memory unit 311 stores one line of pixel data input to the encoding unit 221. This stored 1547
The pixel data for
The color (white or black) and length (run length) of the pixel are detected.

Based on the run length detected by the run length counter 313, the code conversion unit 319 searches the code table 321 in which the terminating code word and makeup code word of -dimensional encoding are stored, and determines the color of the pixel. and a variable length code word corresponding to the run length. The read code word is stored in the memory section 325. By repeating such a procedure, one-dimensional encoding for one line is executed.

Also, when encoding one line, the leading white run length value is used as the white run length value of the reference line in the temporary save register 3.
Store it in 15. When the first pixel of one line is a black run, the white run length value is set to 0.

In encoding the next line, the vertical mode determination unit 317 compares the first white run length value detected by the run length counter 313 with the first white run length value of the reference line stored in the - time save register 315. However, if the difference between the run length values is less than or equal to a predetermined value (for example, 3n@element), the code conversion unit 319 is instructed to perform vertical mode encoding. When the vertical mode is specified, the code conversion unit 319 performs two-dimensional encoding (vertical mode encoding) by searching the code multiple 323 in which the vertical mode code word is stored. Furthermore, if the difference in run length exceeds a predetermined value, the code conversion unit 319 is instructed to perform one-dimensional encoding. After the encoding instruction is given, the leading white run length value stored in the -time save register 315 is updated in preparation for encoding the next line.

Thereafter, the -dimensional encoding process is repeated until the end of one line.

Note that the above-described method of performing vertical mode encoding only on the leading white run length of each line is referred to as a two-dimensional encoding method (FMH encoding method) limited to white pixels at the beginning of a line, and will be described below.

FIG. 4 shows the operational flow of the encoding process of the embodiment. The encoding operation will be explained below with reference to FIGS. 2 to 4.

First, the run length is detected by the run length counter section 313 (step 411). Next, the code conversion unit 319 determines whether or not to perform F M f (encoding).
step 412).

8 If the judgment is negative, perform the normal one-dimensional encoding process shown below.

The code conversion unit 319 adds a line end code EOL to the beginning of each line (step 413), and then determines whether the run length detected in step 411 corresponds to a black run. (Step 414). In the case of a black run, an affirmative judgment is made and a one-dimensional code W(0) is created with the white run length value as 0 (step 415), and one-dimensional encoding of the black run corresponding to the detected run length n ( n) (step 416). Also, if a negative judgment is made in step 414 (in the case of a white run)
creates a one-dimensional code W (n) of a white run corresponding to the detected run length n (step 417).

Also, the determination in step 412 (determination of FMH encoding)
If an affirmative determination is made in step 41, the code conversion unit 319 then determines whether the line of interest is a line or not (step 41).
8). Here, the K line indicates a line of one-dimensional encoding that is forcibly inserted into multiple lines including two-dimensional encoding when encoding by FMH encoding, for example, one line is inserted into four lines. Suppress one-dimensional encoding by line ratio. If a positive determination is made in step 418, skip next (
(described later) and reset the counter for counting the lines (step 419). For example, K
When the counter is set, it is set to a predetermined initial value.The value of the counter is set to a predetermined initial value by the subtraction operation described later.
o'', an affirmative determination is made in step 418.

After that, the code conversion unit 319 creates a line end code EOL+1 to be added to the beginning of the -dimensional encoded line (step 42o). Thereafter, the processing after the determination in step 414 is performed to create a one-dimensional code corresponding to the detected white run or black run.

Further, if a negative determination is made in step 418 (determination of K line), a subtraction operation of the count value of the K counter is performed (step 421). Next, the vertical mode determination unit 317
performs a coding mode determination operation based on the starting white run length of the coding line inputted from the run length counter 313 and the starting white run length of the reference line stored in the -time retreat register 315. (Step 422), the code conversion unit 319 determines whether or not the mode is vertical mode (Step 423). If the difference between the run length values of the reference line and the encoded line exceeds, for example, 3 pixels, encoding in vertical mode is impossible, so a negative determination is made, and the process moves to step 419 described above to perform one-dimensional encoding. Performs encoding processing.

Further, in the case of the vertical mode, the code conversion unit 319 makes an affirmative determination in step 423, and then determines whether or not the respective leading white run length values of the reference line and the encoded line are the same (each leading white run length value is the same). When the vertical mode code corresponding to the difference n in length values is V (n), (2) (determination as to whether or not it is 0) is performed (step 424). If the leading white run length values of each line are different, a negative determination is made, and then it is determined whether or not a skip is being executed (step 42
5). Here, skipping means that if there is a continuation of all white lines, the second and subsequent all white lines will be ■ (0).
This is a process that focuses on the fact that only the code is used and omits the continuation <IEOL 10 to increase the compression rate.

If an affirmative determination is made in step 425, the process moves to step 419. If a negative determination is made in step 425, EOL+0 is added to the beginning of the line to be encoded in the vertical mode (step 42G), and then the vertical mode code V(n
) (n is the difference between the respective run length values of the reference line and the encoded line) (step 427).

Further, if the determination in step 424 is affirmative, the code conversion unit 319 then determines whether or not skip is being executed (step 428), and in the case of an affirmative determination, step 427)
Proceed to processing. In this case, n=D in the vertical mode code V (n).

If skipping is not in progress, a negative determination is made in step 428, and it is determined whether the previous line is completely white (step 429).
. In the G3 facsimile machine, one line represents 1728 pixels, so the starting white run length value of the reference line is 172.
In the case of 8, an affirmative judgment 2 is made and the skiving operation is started and the counter is reset (step 430). Then step 427
Then, the vertical mode code ■(n) is created. In addition, if a negative determination is made in step 429, step 426 (creation of EOL+0) is executed, and then step 4
The process moves on to step 27.

In this way, the white run length at the beginning of each line is encoded.

Next, the code conversion unit 319 stores the above-mentioned white run length value of the encoded line in the temporary save register 315, and in step 431L determines whether or not encoding of one line has been completed (step 432). , the next run length of the corresponding line is detected (step 433).
, it is determined whether the detected run length is for a white run (step 434). In the case of a white run, an affirmative judgment is made and one-dimensional decoding (n) of the white run corresponding to the detected run length n is performed in a step 435).
, and then returns to step 432 for determining whether the l line has ended. In the case of a black run, a negative judgment is made in Step 434, and one-dimensional decoding (n) of the black run corresponding to the detected run length n is created (Step 436), and then judgment is made in Step 432 (one line is completed). Return to ).

When the encoding for one line is completed, the code conversion unit 31
9 makes an affirmative judgment in step 432, and then F M H
It is determined whether or not it is encoded (step 437). If the determination is negative, the necessary number of fills FILL are created (step 438). When the creation of the file FILL is completed, or when an affirmative determination is made in step 437, the code conversion unit 319 then determines whether or not encoding for one page has been completed (step 439). If there are other lines to be encoded, make a negative determination and proceed to step 4.
Repeat the process from step 11 onwards. If a negative determination is made, a control return code RTC is activated (step 440), and the encoding process for one page of the transmission document is terminated.

Ba1yu” 0HU4 luck FIG. 5 does not show the detailed configuration of the decoding section 231.

4 In FIG. 5, 511 and 523 indicate the memory section, and 513
is the code detection part, 515.51'? is the decryption table,
Reference numeral 519 indicates a temporary save register, and 521 indicates a run generation section.

The memory unit 511 stores encoded data received by the facsimile device 210, and the stored encoded data is sequentially read out and input to the code detection unit 513. The code detection unit 513 searches the decoding table 515 in which the terminating codewords and makeup codewords of -dimensional encoding are stored, based on the input encoded data, and reads out the corresponding run length value. . The run generating section 521 is
Pixel data is generated according to the read run length value and output to the memory section 523. By repeating such a procedure, one-dimensional decoding for 1947 parts is performed.

Further, during the -dimensional decoding process, the leading music run length value is stored in the temporary save register 519 as the leading white run length value of the reference line.

In the decoding process of the next line, when the code detection unit 5513 detects the vertical mode code as the first code of one line, it reads the start run length value of the reference line stored in the - time save register 519 and reads the vertical mode code as the first code of one line. A correction is made according to the mode code, and this is used as the leading white run length value. The code detection unit 513 generates a decoding table 5 in which vertical mode code words are stored based on the vertical mode code.
Decryption is performed by searching for 17. If the first codeword is not a vertical mode code, -dimensional decoding processing is performed. Furthermore, after the vertical mode code is detected, one-dimensional decoding is repeated until the end of one line.

Summary of n In this way, by performing vertical encoding by focusing only on the white run length at the beginning of each line, efficiency can be improved by adding a small amount of processing flow compared to the -dimensional encoding method. Good encoding can be performed, and therefore, it is possible to achieve higher communication speeds and lower device costs.

For example, -6 of the vertical mode code used in the example Examples are shown in Table 1.

Table 1 In Table 1, VR is the vertical mode code V (
n), and conversely VL shows the vertical mode code V (n) when the change point of the encoded line is to the left of the change point of the reference line. Further, the numbers in parentheses indicate the difference in change points, that is, the difference n in the leading white run length values.

CC using FMH code encoding using such a vertical code.
An example of encoding an ITT Nα1 chart (the composition ratio of all white lines is assumed to be 50% for simplicity) is shown below (1) Total number of lines per page: 297 (mm)
X3.85 (line/nun) - approx. 1150 lines (2
) Number of all white lines per page: 1150/7 2 = 475 lines (3) Number of consecutive blank lines when K = 8 = 57
5X7/8=503 lines Also, since the white area is separated by 30 character lines,
100 lines are all white starting lines (EO+-+OL 47
5 consecutive all white lines (E OT-omitted), remaining 5 lines
Assuming that 75 lines are normal one-dimensional encoding lines, this 1
/2 (289 lines) allows replacement with a vertical mode code.

As a result of encoding the example for such a CCI T In the case of compression, it is set to 1). This value is almost equivalent to the boundary differential sequential encoding method (MR method), which is a typical two-dimensional encoding method.

Furthermore, in the configuration of the encoding unit 221 shown in FIG.
Memory section 311 325. - Time save register 315. The code tables 321, 323, etc. are stored in the facsimile machine 21.
The memory section 2178 in C.0 can be substituted, and the other components in FIG.
This can be easily realized by software processing of the PU 211. Therefore, there is no need to add particularly complicated hardware when performing the above-mentioned FMH code encoding.

■4B・ノ1i= Note that in the embodiment of the present invention described above, the vertical codes shown in Table 1 are newly defined, but CCITT Recommendation T,
A vertical mode code according to the two-dimensional code table of No. 4 may be used.

In addition, in the embodiment, only the first white run length of each coding line is encoded in the vertical mode, but if there is often a black run at the beginning of each line, the first black run length is It is also possible to encode only the vertical mode.

Furthermore, in "1. Correspondence between Examples and FIG. 1",
Although the correspondence between the present invention and the examples has been explained, those skilled in the art can easily imagine that the present invention is not limited to this and that there are various modifications. Dew.

〔Effect of the invention〕

As described above, according to the present invention, by performing two-dimensional encoding on the first pixel change position of each encoding line, it is possible to increase the efficiency of encoding and achieve faster communication speed. At the same time, by performing two-dimensional encoding only on the first changed pixel position of each encoded line, the processing is simplified and can be realized by rough 1-ware processing.
It is extremely useful in practical terms.

[Brief explanation of drawings]

FIG. 1 is a principle block diagram of the encoding method of the facsimile machine of the present invention, FIG. 2 is an overall block diagram of an embodiment of the present invention, FIG. 3 is a block diagram of the encoding section of an embodiment, and FIG. FIG. 4 is an explanatory diagram of the operation of encoding processing in one embodiment, and FIG. 5 is a diagram showing the constituent countries of the decoding section in one embodiment. 0, 111 is a detection means, 121 is a storage means, 131 is an encoding determination means, 141 is an encoding means, 210, 290 is a facsimile device, and 211 is a CPU. 213 is a reading unit, 215 is a recording unit, 217, 311, 325, 511, 523 is a memory unit, 219 is an operation panel unit, 221 is an encoding unit, 231 is a decoding unit, 241 is a modem, 243 is a line control unit , 313 is a run length counter, 315.519 is a temporary save register, 317 is a vertical mode determination section, 319 is a code conversion section, 1 321 323 is a code table, 513 is a code detection section, 515.517 is a decoding table, 521 is a code conversion section. This is the run generating area. 2 Decoding 4th p 5th

Claims (1)

[Claims] (1) Detection means (111) for detecting the changed pixel position for each encoded line of the transmission document, and storage means (111) for storing the first changed pixel position of the reference line.
121), the first changed pixel position of each encoded line detected by the detection means (111) and the value stored in the storage means (121) are compared to determine whether two-dimensional encoding is possible. The changed pixel position detected by the encoding determining means (131) and the detecting means (111) are input, and if two-dimensional encoding is possible for the first changed pixel position of each encoding line, two-dimensional encoding is performed. encoding means (141) for performing dimensional encoding and performing one-dimensional encoding for other changed pixel positions of this encoded line; An encoding method for a facsimile machine, characterized in that it is configured to perform dimensional encoding.