JPS6370671A - Document communication method - Google Patents

Abstract

PURPOSE:To simplify protocol and to improve transmission efficiency by dividing data into plural blocks when two kinds of data are transmitted and performing transmission by redividing processing when the data are divided into blocks more than the specific number of blocks. CONSTITUTION:The code block 12-1 of an overlap block 12 is developed into bit image data in a program memory 23 through CG. This data and bit image data of a bit image block 12-2 are put together by overlaying and stored in an image memory 25 in image pattern like (C). In such a case, the border and attribute of a code block 12-1 are erased, so only three blocks 11 - 13 are required and the transmission efficiency is high. Further, the border and attribute of the block 12 are the same with the block 12-2 of the bit image. The borders of blocks 11 - 13 are included in the headers of the respective blocks.

Description

[Detailed Description of the Invention] [Technical Field] The present invention relates to document communication, particularly mixed data documents [Prior art] Mixed terminal transmission using a method that allows data of different types such as character code data and bit image data to be mixed and transmitted. As a method, a method has been considered in which one text (1) is assembled into a set of multiple blocks including character code blocks and bit image blocks, and the blocks are sent sequentially.

However, when a large number of text, photographs, and drawings are mixed together, the number of blocks described above becomes large, and the protocol for transmitting the blocks becomes complicated, and it takes time to assemble the original document on the receiving side. There was also the disadvantage that transmission efficiency deteriorated.

〔the purpose〕

The present invention provides a mixed data communication method that eliminates the above drawbacks, and the present invention provides a mixed data communication method with high transmission efficiency. The present invention also provides a communication method that can easily reproduce an original document from mixed data. The present invention also relates to a data transmission method that divides a document into small blocks and transmits the document.

The present invention resides in a transmission system that can mix and transmit data based on image processing, and a communication terminal device that enables efficient processing of mixed data.The present invention also resides in a transmission method that can transmit data based on character recognition.

The above objectives and other objectives will become clear from the following examples.

〔Example〕

FIG. 1 is a block diagram showing one embodiment of the present invention, and FIG. 2 is a perspective view showing the above embodiment.

The reader 10 reads a predetermined document and outputs an electrical signal.

The facsimile body 20 includes a reader/printer interface 21, an image compression unit (hereinafter referred to as rlcUJ) 22, and a program memory (hereinafter referred to as rlcUJ).
PMEMJ) 23, bit move unit (hereinafter referred to as BMUJ) 24, and image memory (hereinafter referred to as rI)
25 (hereinafter referred to as "MEM"), and video RAM (hereinafter referred to as rVRA).
(hereinafter referred to as MJ) 26, a central processing layer (hereinafter referred to as rCPUJ) 27, and a communication interface 28.
, a bus 29, and a communication control unit (hereinafter referred to as "ccU") 30.

The ICU 22 compresses or expands data,
Two-dimensional compression (high compression) is used to increase the coding rate.

The PME"M 23 has an OS program and an application program memory area for controlling the input/output devices installed around the facsimile main body 2o and each unit in the facsimile main body, and also stores character code data as an image. It has a font memory area for converting into bit data.It also has an area for storing and editing text code data (character data) by keystrokes or word processing.

In addition, PMEM23 is a memory management unit (
It also has a work area as a buffer for transmission data for transmitting data from the hard disk 50 via the CCU 3O and storing data from the CCU 3O to the hard disk 50. Note that the buffer described above is for adjusting the speed of the disk, line, etc.

The BMU 24 processes data bit by bit in the CRT 60 and performs image editing (image processing), and enlarges, reduces, rotates, moves, or extracts a predetermined image.

The IMEM 25 has 4 Mbytes, and can store image data from the reader, edited image data by the BMU 24, data decompressed by the ICU 22, character code data manually created, and text created by word processing. Bit data obtained by converting code data, mixed data, or character code data into an image (for example, expressed as 1 pixel and 1 bit)
It is used to store. Here, the mixed data is 1
Image bit data and character code data coexist within the page. Each block is managed and stored with an identification code called an image block or a character block. In addition, the IMEM 25 temporarily stores predetermined data so that the reader 10, printer 7
0. It may also be used as a buffer to match the speed of line 40.

The VRAM 26 stores image data to be displayed on the CRT 60 in the form of bitmap data (for example, one pixel corresponds to one bit).

Furthermore, a hard disk device 50 and a floppy disk device 51 are provided as external storage devices. Although these devices are non-volatile memories, backup memory may also be used as the non-volatile memory. Transmitted data and received data are stored and saved.

The keyboard 61 is used to store command data for sending and receiving commands, command data for image processing, word processing, etc.
Manually generate character data for word processing.

The pointing device 62 moves a cursor image on the CRT 60 and specifies a position for image editing or the like using the position of the cursor image. This 62 also divides the mixed data into blocks. The coordinates indicating this block are stored and managed by the PMEM, and are used as one piece of data in identification code data (header) at the time of transmission.

The line 40 is preferably a digital data conversion network or a digital data packet network, such as 64 bits/se.
C digital lines are convenient for high-speed, large-volume transmission of image data with a large amount of information and high resolution (density).

The printer 70 is a 3Mbp laser beam printer.
Print at a speed of s.

Figure 3 shows one page of mixed data divided into blocks 1 to 8. This one page data corresponds to one page display on the CRT 60 screen, and also corresponds to one page data at the time of transmission. It also corresponds to one page of the print sheet printed at the time of reception. Also, create multiple pages of such page data, store them on the hard disk 50, and store the multiple pages in one page.
As shown in Figure 4, the transmission arrangement of the mixed data that is transmitted each time is such that the structure data (header 1) is placed before the block data 1 to n.
~n) is added.

This header indicates an identification signal indicating whether the next block data is image data or character data, the size (data amount), and position on one page of the block data. A is an acknowledgment signal for sending and receiving block data.

When one page has been transmitted, an EOP is transmitted.

The data of one page of the original from the reader 10 is stored in the image memory 25, transferred to the VRAM 26, and read in the CR.
That one page is displayed on T60. The image is trimmed via the BMU 24 in response to editing instructions from the keyboard 61 or point device 62, and only the image corresponding to block 3 is stored in the memory 25 again.

Next, the text code data from key 61 is stored in memory 2.
3, is bit-converted, and transferred to the VRAM 26, where one page of text is displayed on the CRT 60. The information is edited in the same way as image processing, and blocks 1 to 1 in Figure 3 are edited.
6 corresponding text characters and stored in the memory 23 again. In this case, the data stored in the memory 23 is a code. Note that the position data corresponding to each block is managed as an attribute code in the memory 23 together with the data type. With the next command, the image data and position data of block 7.8 of the memory 25 are read out and transferred to the memory 25 via the VRAM.
It is displayed at the position of block 3 of the RT 60, and finally blocks 1 to 8 in FIG. 3 are displayed. This mixed data is read from each memory 23, 25 in the order of blocks 1 to 8 and stored in the disk 50 in order.

FIG. 3(C) shows a one-page document as a result of editing. Blocks 1 to 6 are character code blocks, and blocks 7 and 8 are bit image blocks.

In this way, when character codes and bit image blocks coexist in a document, a large number of transmission blocks are required.

This is divided into two blocks in Figure 3 (a) and (b).
). They become blocks 9 and 1o, respectively.

In FIG. 3, (a) is a code block, and since the part corresponding to block 7 in the code block is the end of the line, there is no need to fill in any code. In the area corresponding to block 8, the space between the next character code and the next character code is filled with a blank code such as a space or a tab.

Figure 3(b) is a bit image block, and in order to make it the smallest rectangular block, it must be one block that includes at least blocks 7 and 8 as shown in the figure (there is no particular restriction on this. (optional). Figure 3 (C
) is filled with all white bits.

FIG. 5 shows a processing chart of block conversion.

The bit image data from the reader 10 and the code data from the keyboard 61 are combined on the CRT 60 in the process described above to edit the document as shown in FIG. Enter the code indicating the coordinates and block size) and the block attributes. It is determined from the attribute code whether there is a character code block in it 2), and if it is, it is determined whether it is multiple blocks or not 3), and if there are multiple blocks, the block 8 in the image area between the blocks is determined by the space code. Fill in (4). Block 7 has a character line return code at the border with block 2, so there is no need to fill in the space code.

Then, delete the codes that indicate the block boundaries, change the attributes, and create PMEM as one block with only character codes.
(5).

Next, determine whether there is a bit image block from the attribute code 6), and if so, determine whether there are multiple bit image blocks 7)
), if there are multiple, fill in between blocks with all white bits (8
). Then, the code indicating the block boundary is deleted, the attributes are changed, and the block is stored in IMEM as a single block of bit images (9). Next, we need to determine whether there are multiple blocks in total (10). Since there are currently two blocks, character code block 9 and image block 10, this means that these two blocks are superimposed on each other based on point A of block 9. Set the superposition attribute code to PMEM. Then, by inputting a transmission command, the superimposed attribute code is transmitted, and then the code data and bit image data stored in PMEM and IMEM are sequentially transmitted. The overlapping attribute code can also be sent after sending the document data.

On the receiving side, the transmitted block data 9.10 is stored in the disk 50, and then PMEM.

The data in block 9 is converted into bit image data by a character generator, and the encoded data in block 10 is converted into bit image data by a decoder. By decoding the data into image data and superimposing the data corresponding to each block on the point A reference, the transmitted text shown in FIG. 3(C) can be reproduced on the CRT 60 or printer 70.

FIG. 6(C) is a document having a block 12 in which a character code and bit image data are superimposed between character code blocks 11 and 13.
can be divided into a code block 12-1 and a bit image block 12-2 of the same size and sent. In that case, there will be a four-block configuration, but the number of blocks will be fewer than if the superimposed block 12 is divided into smaller blocks (further divided into overlapping blocks and non-overlapping blocks), and the transmission efficiency will be improved.

In this case, a superimposed attribute code is given to the block 12 while leaving the block boundaries and attributes intact, and the receiving side reproduces only the transmitted blocks 12-1 and 12-2 in a superimposed manner.

In addition, Figures 6(a) and 6(b) include overlapping blocks.
4. It is composed of bit image blocks 15),
In the same manner as described above, block boundaries are deleted, a B-point-based overlapping attribute code is assigned, and the data is transmitted, and the receiving side combines the data of blocks 14 and 15 according to the attribute code and displays or prints it.

Further, the code block 12-1 of the superimposition block 12 is developed into data as a bit image via CG in the memory PMEM, and this data and the bit image block 12-2 are expanded.
The image pattern shown in (C) is created by overlay synthesis (taking logical OR) with the bit image data of
It can also be stored in MEM. In this case, since the boundary and attributes of code block 12-1 are deleted, only three blocks (11, 12, 13) are required, and the transmission efficiency is increased in this respect. The boundaries and attributes of block 12 are the same as block 12-2 of the bit image. In addition, block 11~
13 boundaries (positions) and attribute codes are included in the labels C and I of each block in FIG.

Next, optimal division of transmission blocks will be explained. 7th-
1 to 7-4 are examples of possible region division of a one-page text having a character code data region C and an image data region I.

In the example shown in FIG. 7-1, the character area is c, . . . C2, the image area is 11, and transparent superimposition is not performed.

In this case, the transmission format is C1 block, C2 block, ■
This is a three-block transmission method with one block.

Note that the starting point and size of each block are determined by inputting the coordinates of P, , P2.

In FIG. 7-2, the character area 01' and the image area 11 are treated as transparent overlapping. In C8', the area outside C3 is an area without character code data, and contains the return code at the end of the character line of CI. In this case (:+', C
This is a 3-block transmission of 2 and 11.

In Figure 7-3, character areas C, , C2 are unified and 1 of CA is
It is configured as a page and transparently overlaps with image area I. In this case, one page of CA and one block of I are transmitted.

In FIG. 7-4, character areas C, , C, and image area (1) are unified (CA and IA) and transparently superimposed.

In this case, two pages will be transmitted.

Although it is not univocally determined which transmission efficiency is higher, in this example, the character area occupies a considerable area, so the 7th
It is thought that the transmission of one page of character codes and the transmission of one block of images as shown in Figure 3 will be more efficient. This is because the Herada code for the CI and C2 blocks is no longer necessary.

Also, in Fig. 7-4, the code for each block is not required, but during image creation IA, areas other than 11 require uniform bit data equivalent to white or black as image data, and even if compressed and encoded. will also increase the amount of information. Therefore, this case is considered disadvantageous.

However, as shown in Figure 7-5, when the two blocks I, I2 occupy most of one page, two-page transmission is performed (II, I2 are grouped into groups 1 and C, C, ~C3) as shown in Figure 7-4. group 2) may be advantageous.

In this way, the appropriateness of region division differs depending on the type and distribution of information.

In order to determine this appropriate division, the total data amount of communication information at each division is calculated, each is compared, the division procedure with the smallest amount is selected, and transmission is performed based on the division.

FIG. 8 is a flowchart of the determination control.

First, in S-0, 4 is set as the number n of data divisions for one page of one CRT screen in the case of FIG. S-
The area shown in FIG. 7-1 is divided at t. This is based on the position data of Pl and P2 when editing the document mentioned above.
The code data corresponding to C1 block in M and the code data corresponding to 02 are each divided into data amount C1m.

C2m is calculated. This amount of data is determined by the fact that the memory address up to the P2 position is stored in PMEM in advance.
This will be required from now on. At S-2, the bit data corresponding to block 1 in the memory IMEM is compressed by the ICU 22, and the compressed data amount is 1. m is determined and stored by the ICU 22, and the total amount M1 of communication data of C, m+c2m+I,m is determined, and F! Stored in MEM. The compressed data is temporarily stored on the hard disk 50.

Returning again to Sl, the area is divided as shown in FIG. 7-2, and the amount of data corresponding to block 01' is determined. In this case, there is almost no difference from C1 in FIG. 7-1.

Let the total amount be M2.

Next, the data amount corresponding to block CA is obtained by dividing as shown in FIG. 7-3. The total amount is M3.

Next, the division shown in FIG. 7-4 is performed to obtain the amount of data corresponding to block IA. This block has a compressed data amount of white bits other than those corresponding to 11. It is added to m. Let the total amount be M4.

In the case of division, the number n is set to 4 in advance, so in S-3, n is subtracted by 1 for each total amount calculation, in S-4 it is determined whether or not O has been reached, and in S-5, The respective total amounts M and -M4 in the case of four divisions are compared.

As a result, the division mode with the minimum amount of data is determined.

According to the determined division mode, one page of data is divided into areas and stored on the hard disk, and the divided blocks are sequentially transmitted in response to a transmission command.

Above, we determined the appropriate division mode based on the total amount of data, but if the total amount of data does not change much as shown in Figures 7-2 and 7-3, it is better to divide it into two transmissions rather than three transmissions. If it is better to send the data in terms of transmission efficiency, it is preferable to determine the appropriate mode depending on the number of divided blocks. The number of blocks is stored for each division shown in FIGS. 7-1 to 7-4.

Note that it is also possible to arbitrarily select one of the division modes by manually specifying the block division.

In the above, the appropriate division mode can be determined by a transmission command and the transmission operation can be performed automatically according to the completion of the division, or the various divisions can be executed by a preliminary command, the appropriate division can be displayed, and the transmission can be performed by a subsequent transmission command. It is possible.

As an example of culture, 6 Normally, block division processing and transmission as shown in standard Figure 7-1 is performed, but if the number of block divisions exceeds a maximum predetermined number (for example, 31), it will be less than that number. It is also possible to force selection of various division modes. In this case, as shown in Figure 8-2, it is determined whether the number of blocks is less than or equal to the maximum number MAX, and only when the number of blocks exceeds the maximum number, the flowchart in Figure 8-1 is executed, and the maximum number of blocks is determined based on the information amount of each division mode. to determine the optimal mode below. This saves time required for pre-processing before transmission.

Next, an example in which a character recognition function is added to the above-described embodiment will be described below.

In the case of this embodiment as well, the configuration is the same as that shown in FIG. It will be done. This character recognition method is performed using a generally well-known method.

When recognizing character information on a document, there may be characters that cannot be recognized. Therefore, if unrecognized characters are sent as separate blocks as image data, block division becomes complicated and the number of divided blocks increases.

Therefore, in this embodiment, recognized characters are transmitted as code blocks as shown in the example of FIG. 3, and unrecognized characters are transmitted as bit image blocks, which are superimposed and synthesized on the receiving side.

FIG. 9 is a diagram showing an example according to the fourth embodiment.

When character recognition is performed on the manuscript in Figure 9(a), "terminal", "
Characteristics” characters are unrecognizable.

In the code block of FIG. 9(b), blank codes are assigned to unrecognizable characters.

FIG. 10 is a flowchart showing the control operation of the CPU 27 in the fourth embodiment.

The fourth embodiment will be described in detail below based on the flowchart shown in FIG.

In step S1 in FIG. 10, the CPU 27 stores the bit image data of the document read by the reader 10 in the IMEM 25.

Then, in steps 82 to S6, character recognition is performed one character at a time. Regarding this character recognition, first, a character line is recognized by scanning the bit image data of the IMEM 25, and when the character line recognition is completed, a character string is next recognized. In this way, the bit image data of the original is a predetermined character line.

It is divided into character strings, and each character is recognized one by one.

There are various character recognition methods, and the method is not limited to the method of this embodiment.

One character is recognized in steps S2 and S3, and when the character is recognized, the corresponding character code PME is determined in step S4.
Store it in the code block area of M23, and if the character cannot be recognized, store a space code in the code block area of PMEM23, and write the bit image data of the unrecognized character to the address of the unrecognized character in the bit image creation area of PMEM23. Store. Here PMEM
To explain each area of 23, PMEM 23 is provided with a codebook area, a bit image creation area, and a bit image block area. The code block area and the bit image creation area are each given addresses for the number of divisions (character rows x character strings) for character recognition. Further, the size information (a, b) of the character bit image block is given to the image for one character to be recognized as shown in FIG. 9(C).

Next, when it is determined in step S6 that recognition of all characters has been completed, the process proceeds to step S7, and in step S7 it is determined whether or not there are any unrecognizable characters.

If it is determined in step S7 that there are no unrecognized characters, step S
Proceed to step 8 and write the address information and size information of the code block before the data in the code block area of PMEM23.
For example, (x, y) and (x', y') in Figure 9(b)
, and information indicating the arrangement of characters and identification data indicating that it is a code block are added and transmitted as code block data.

If it is determined in step S7 that there is an unrecognized character, then in steps S9 to S12, the minimum Xmin=X+ and the minimum Ymin=y+ of the address where the bit image data of the unrecognized character in the bit image creation area of the PMEM 23 is stored.

Maximum Xmax = xn, maximum dog Ymax = maximum (Fig. 9 (
C) is read out, and a bit image block is created in steps 513 to S15. First step S
13, bit image block starting point address information (X
min, Ymin) and size information (Xmax+a-
Xmin, Ymax+b-Ymin), and in step S14, identification information indicating that the block is bit image data and block overlapping attributes indicating that the block is overlapped with the code block are set. Then, in step S15, the bit image data of the bit image creation area is stored in each area where the unrecognizable character is located in the bit image block area, and the bit image data as shown in the bit image block of FIG. 9(C) is created. Created.

The bit image data thus created in step S15 is processed as bit image block data in step S15.
13. It is stored in the PMEM 23 together with the information set in S14. At this time, the bit image data is
It may also be encoded by .22.

Next, in step S16, address information and size information of the code block are first added to the code data in the code block area, as in step S8.

After adding information indicating the arrangement of one character of identification information and storing it in the PMEM 23 as code block data, the code block and bit image block are transmitted to the communication partner.

As described above, according to the fourth embodiment, since the characters of the original are recognized and the characters are encoded, the effort required for operation can be saved compared to inputting characters using the keyboard 61, and the data communication time can be shortened.

Moreover, since characters that cannot be recognized are transmitted as bit image data, the original data can be reliably transmitted.

Furthermore, since recognized characters and unrecognized characters are each included in one code block and one beat image block and transmitted, the number of blocks is smaller than when data is divided into multiple blocks, and the data Communication time is shortened, and data processing on the sending and receiving sides is also simplified. It is also possible to include code data of a text created by a word processor in the code block obtained by the recognition described above and send it as one block.

The different data include graphic code data, character code data, line image bit data, halftone image data, and the like.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a communication terminal device according to the communication method according to the present invention, FIG. 2 is a perspective view of a communication terminal device according to the present invention, FIG.
Figures 4, 6, 7-1 to 7-5, and 9 are diagrams showing data formats, Figure 5, Figure 8-1, Figure 8-2, and Figure 10. is a flowchart diagram showing a communication method.

Claims (1)

[Claims] A document communication method for transmitting a mixture of two or more types of data, in which the data is divided into multiple blocks, and when the number of divisions is greater than a predetermined number of blocks, the number of divisions is reduced to less than the predetermined number of blocks and transmitted to a re-division process. .