JPH0522574A - Binary picture data segment system - Google Patents

Abstract

PURPOSE:To make it possible to perform the partial segment of a picture and a reencoding processing after an editing using a small memory at a high speed even if the picture is one from a large-sized binary picture. CONSTITUTION:A binary picture compression/extension processor 10 encodes a starting line by a code word having no correlation in a vertical direction by the instruction of a horizontal mode instruction circuit 17, stores the other lines as the code data encoded by a prescribed code word and stores code position data showing the starting position of the code word corresponding to the starting line as an intermediate start table, for plural areas (partial image) obtained by dividing all the pictures every prescribed number of line. At the time of segmenting the partial messages, an extension processing only for the necessary partial image is performed based on the code data and encoding position data. At the time of also encoding the image for which an editing, etc., was performed, a compression processing is performed only for the partial image and the image is connected with unedited code data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a binary image data cutout system which is applied in a binary image compression / expansion processing device or the like and which cuts out a part of large size image data.

[0002]

2. Description of the Related Art In a binary image compression / expansion processing apparatus which handles a large amount of binary images, binary images are often coded and handled in order to speed up data transfer. An MH (Modi
fied Huffman) method, MR (Modified READ) method, M
There is an MR (Modified MR) method.

Of these, the MMR method has the highest compression rate and is often used in image processing systems and the like. The MMR code is a method of encoding by utilizing the correlation between the line immediately before and the vertical direction for scanning in the line direction. When decompressing an MMR code, since there is no code (EOL code) indicating a line break, it is necessary to sequentially decompress the image data of the reference line from the leading code.

Therefore, when the MMR-encoded binary data is partially cut out, edited, etc., and then compressed again to the encoded data, conventionally, the entire image is decompressed into bit map data and the necessary part is decompressed. The image was cut out, the data was changed according to the edited contents, and all bitmap images including the cut out image were encoded.

[0005]

When this method is used, a memory for all bit map images is required, so that when the image size becomes large, a large capacity memory is required accordingly.

Further, since all the bit map images are first decoded, the images which are not necessary for editing are also decoded, which requires a long processing time. At the time of encoding, since the image portion that has not been changed is also encoded again, it takes time to process.

The present invention has been made in view of the above points, and uses a small memory for partial cutting of an image and re-encoding after editing even from a large size binary image. It is an object of the present invention to provide a binary image data cutting method that can be performed at high speed.

[0008]

According to the present invention, in a system for compressing / decompressing a binary image, a start line is vertically set for each of a plurality of regions obtained by dividing the binary image by a predetermined number of lines. Coding means for coding with a code word having no correlation in the direction and coding other lines with a predetermined code word; and first storing means for storing the code word obtained by the coding means, A second storage unit for storing code position data indicating a start position of a codeword corresponding to each start line of the area in the codeword stored by the first storage unit; and the second storage unit. An image cutting-out unit that cuts out a partial image in the binary image for each area based on the stored code position data and the code word stored in the first storage unit. To do.

Further, the present invention is a system for performing compression / expansion and editing of a binary image, wherein the binary image is encoded for each of a plurality of regions obtained by dividing the binary image by a predetermined number of lines. Means, a codeword storage means for storing the codeword obtained by the first encoding means, and a code corresponding to each start line of the area in the codeword stored by the codeword storage means Position storage means for storing code position data indicating a start position of a word, code position data stored in the position storage means, and the binary based on the code word stored in the code word storage means Image cutting means for cutting out a partial image in the image for each area, and second coding means for coding the binary image cut out and edited by the image cutting means for each area. , The second encoding A code word combining means for connecting the code word obtained by the step with the code word corresponding to the region not cut out by the image cutting means and storing it in the code word storage means. is there.

[0010]

According to such a configuration, when a partial image is cut out from a large size binary image, only a partial image (area) including a necessary portion can be an object of decompression processing. That is, the head position (code position data) of the code stored in the position storage means (intermediate start table) can be referred to and decoding can be performed only from the necessary area. At that time, since the leading line of each partial image is stored by a code word having no correlation in the vertical direction (horizontal mode code word),
It can be expanded regardless of the reference line data. Therefore, general decompression processing is possible after the first line.

Further, after the cut-out image is edited, when it is re-encoded, even if the code length differs from the original code length, it is stored by being connected to the code word corresponding to the non-cut-out area. It The position storage means stores the code position data indicating the start position of the code word corresponding to each start line of the area. Therefore, when the code length is changed by editing the image, the code position data is changed accordingly. To be done. Therefore, partial image cutting and editing can be repeatedly performed thereafter.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. Here, a case where the MMR coding method is used will be described as an example. FIG. 2 is a block diagram showing the configuration of a system to which the binary image data cutout method according to the embodiment is applied. As shown in FIG. 2, this system has a CPU.
2, a main memory 4, a display device 6, a mass storage device 8, and a binary image compression / expansion processing device 10 are connected via a system bus 28. The main memory 4 stores image data cut out as an object to be edited, and is rewritten according to the content of editing. The display device 6 displays an image to be edited when performing image editing or the like. The mass storage device 8 is composed of a magnetic disk device, an optical disk device, or the like, and stores code data and the like obtained by compressing image data by the binary image compression / expansion processing device 10. Binary image compression / expansion processing device 1
0 is code data (for example, M
A compression process for storing as an MR code) and a decompression process for decoding the code data (MMR code) and returning it to image data are performed.

FIG. 1 is a block diagram showing the detailed arrangement of a binary image compression / expansion processing apparatus 10 in the system shown in FIG. As shown in FIG. 1, the binary image compression / expansion processing device 10 includes a change point detection circuit 12, a decoding circuit 14, a generation circuit 16, a horizontal mode instruction circuit 17, and an I / O register group 1.
8, a byte counter 20, a line counter 22, pointers 24 and 26, and a reference line image memory 27, which are connected to the CPU 2 via a system bus 28.

This system is provided with a large-capacity storage device 8 for storing code data given to the binary image compression / expansion processing device 10 at the time of expansion processing, information as an intermediate start table 50 described later, and the like. ing. The image data to be compressed is, for example, ITV.
It is assumed that the image is input to the system by an image input device (not shown) such as.

The change point detection circuit 12 detects the color change point (b1 point) from the image data of the line (reference line) immediately before the line to be decoded or encoded stored in the reference line image memory 27. It is to detect.

The decoding circuit 14 inputs a code (code data) at the time of expansion processing and inputs image data (image data) at the time of compression processing to decode the data. For example, in the vertical mode of the MMR code, the decoding circuit 14 makes the relative change point position δ (“the change point on the right side of the start point (a0 point) of the encoding line) point b1 and point a1 during the expansion processing. a1-b1 "), the position of point a1 is detected and output during compression processing. Details of the decoding circuit 14 will be described later.

The generation circuit 16 generates image data at the time of expansion processing based on the change point (b1 point) detected by the change point detection circuit 12 and the result (output data) decoded by the decoding circuit 14, Code data is generated during compression processing. Details of the generation circuit 16 will be described later.

The horizontal mode instruction circuit 17 outputs a horizontal mode instruction signal when compressing the first line of each partial image (described later) included in the entire binary image, and a horizontal mode code word (code). It is for executing the encoding in.

The I / O register group 18 holds various data used for compression / expansion processing in the apparatus, and includes therein the number of lines n for dividing the entire binary image into a plurality of partial images. It includes a register for instructing.

The byte counter 20 counts the number of bytes of the code data from the first byte of the code data during the expansion process. The pointer 24 indicates the head bit position of the code at which the decoding circuit 14 should start decoding the code data during the expansion processing.

The line counter 22 counts up to the number of lines n in the partial image set in the I / O register group 18 every time the compression / expansion process is performed for one line.

The pointer 26 indicates the head bit position at which the code should be connected to the generation circuit 16 during the compression processing (including the compression processing after the editing of the partial image).

Next, the details of the decoding circuit 14 will be described. FIG. 3 is a block diagram showing the configuration of the decoding circuit 14. The register 14a is for holding 16-bit input data to be processed. Register 1
Reference numeral 4a denotes a lower 8-bit register C1 for inputting and holding 8 bits of input data, and when 8 bits are input to the register C1, inputs and holds the contents held in the register C1 up to that point. It consists of the upper 8-bit register C0 for The register 14a holds a code / bit string (code data) forming a variable length code string at the time of decoding, and holds image data at the time of encoding.

The funnel shifter 14b includes a register 14a.
It is for taking out a predetermined bit string from the data held in. The pointer 24 indicates the head position of the bit string to be taken out by the funnel shifter 12. For example, MM
When processing the R code, it indicates the beginning of the code during decompression processing and indicates the position of point a0 during compression processing. D-
The PLA 14c is a decryption PLA (Programmable Logic A).
rrey), and a table storing data to be output corresponding to the output of the funnel shifter 14a is provided. The table provided in the D-PLA 14c is provided with a decoding table for one-dimensional codes, two-dimensional codes (MR system, MMR system), special codes and the like. In addition, the code length used for updating the value of the pointer 24 at the time of decompression processing to the next decoding position is stored.

Next, details of the generation circuit 16 will be described. FIG. 4 is a block diagram showing the configuration of the generation circuit 16. An arithmetic processing circuit (adder) not shown is provided in the preceding stage of the circuit shown in FIG. The arithmetic processing circuit
In the vertical mode of the MMR encoding method, at the time of decompression processing, the a1 point (the length of the image to be generated is determined based on the b1 point output from the change point detection circuit 12 and the δ (and a0 point) from the decoding circuit 12). “B1 + a0 + δ” shown is obtained, and the b1 point output from the change point detection circuit 12 and the decoding circuit 12 during compression processing.
Δ (indicating the address of the code ROM 16a described later) is obtained based on the a1 point from

The code ROM 16a shown in FIG. 4 is a coding RO for storing code data and its code length L.
It is M. The code length L is stored in the pointer 26 during compression processing.
It is used to update the value indicated by. Code ROM16
In a, the corresponding code data is read with δ in the vertical mode of the two-dimensional code, pixel column H in the horizontal mode, pixel column P in the pass mode, and run length of the one-dimensional code as addresses. The selector 16b is a code ROM 3 at the time of compression.
Code data from 1 (EROM), a predetermined bit string (8-bit "0 ... 0" or "1 ... 1") at the time of expansion,
Alternatively, code data (SHT) from the funnel shifter 14b at the time of joining the codes after the partial image is edited.
D) is output to the rotate shifter 16c having an 8-bit length. The rotate shifter 16c has a length of 8 bits, for example, and ring-shifts the data input via the selector 16b in accordance with the value of the pointer 26. Pointer 2
Reference numeral 6 is a code assortment pointer which indicates a bit position for assembling bits for the rotate shifter 16c and the selector 16d to generate output data. This pointer 26 is set as a pointer value for the next machine cycle with the image length output from an arithmetic processing circuit (not shown) as a0 point during decompression processing, and the code length L obtained from the code ROM 16a is added during compression processing. Is set as a pointer value for the next machine cycle. Based on the value set in the pointer 26, the register 16e connects the generated image data during decompression and the code data output from the code ROM 16a during compression. RODT07-00
Is output as data for which connection has been completed.
The selector 16f selectively outputs, for example, upper 8 bits and lower 8 bits of the data held in the register 16e. When the data whose generation has been completed in the immediately preceding generation processing step has a length of 1 byte or more, RODT15-08
Is selected and the length is less than 1 byte, RODT07-
00 is selected.

Next, the data structure in the embodiment will be described. FIG. 5 shows a conceptual diagram of the image data 30 corresponding to the original image. The image data 30 shown in FIG.
The conceptual diagram of the two-dimensionally arranged code data 40 corresponding to is shown. Further, FIG. 7 shows a conceptual diagram of the intermediate start table 50, which is generated at the time of image compression or expansion processing and is referred to at the time of performing partial image cutout.

The image data 30 shown in FIG. 6 includes a plurality of partial images 30-1 for each predetermined number of lines (n) for the entire image.
30-2, ...

The code data 40 shown in FIG. 7 is shown in association with the image data 30 shown in FIG. In general, the coded data has different code lengths depending on the content of the image data even if the image data of the same length is coded. Therefore, on the coded data, the divided areas 40-1, 40-
There is no regularity in the start positions of 2, ... (Start positions 0, 1, 2, ...). Further, the code data 40 includes a code (MMR code) coded in the horizontal mode corresponding to the head line of each partial image 30-1, 30-2, ....

In the intermediate start table 50 shown in FIG. 7, each partial image 30 in the image data 30 shown in FIG.
-1, 30-2, ... For each divided area 40 shown in FIG.
The byte position from the beginning of the code data 40 indicating the start position (code start position) of 1, 40-2, ... And the bit position within the byte are stored in association with each other.

In the intermediate start table 50, the partial images in the image data 30 and the code data corresponding to each are made to correspond one to one. Therefore, when the image is partially cut out from the command data 40, the image data 3
If the partial image including the partial image to be cut out at 0 is known, the code start position of the corresponding divided area in the code data 40 is obtained by referring to the intermediate start table 50, and decoding is started from that address to determine the required number of lines. After the decoding is completed, the decoding can be stopped to efficiently cut out the necessary data.

Next, the operation of the embodiment will be described.
It should be noted that here, a process using a two-dimensional code (MMR encoding method) will be described as an example.

First, the intermediate start table 5 shown in FIG.
A method of creating 0 will be described. First, the I / O register group 18 is accessed to set the number n of lines in the partial image in the image data 30. The intermediate start table 50 is created during the first compression processing when the original data is image data.

Image data is a binary image compression / expansion processing device 1
When it is input to 0 (the decoding circuit 14), the compression processing is sequentially executed. At this time, when compressing the leading line of the partial image 30-1, the horizontal mode instruction circuit 17
Instruct to encode the MMR code in the horizontal mode. When the process for the next line and thereafter is started, the encoding is performed by the code word in the predetermined mode. As a result, the code data 40 corresponding to the image data 30 is sequentially generated. The line counter 22 counts the number of lines each time the processing for one line is executed. In addition, the byte counter 20 counts the number of bytes of the created code data.

When the count value of the line counter 22 reaches a value indicating that the preset number of lines (n) has been processed, the binary image compression / expansion processing apparatus 10 causes the CP
Generate U interrupt. At this time, the content of the pointer 26 indicates the leading bit position at which the code of the next line should be connected.

Here, the encoding process is interrupted by an I / O instruction from the CPU 2, and the byte position of the command data (the count number of the byte counter 20) and the bit position (the position where the next code shown by the pointer 26 is connected). Are read in via the system bus 28. On the system side, the data read from the binary image compression / expansion processing device 10 is sequentially stored as an intermediate start table 50.

After that, the encoding process in the binary image compression / expansion processing device 10 is restarted. Thereafter, the above-described processing is sequentially performed for each partial image, whereby the code data 40 of the entire image in which the leading line of each partial image is encoded by the code word in the horizontal mode and the intermediate start table 50 are created.

Next, a specific operation of the binary image compression / expansion processing apparatus 10 for performing partial expansion processing (partial image cutting) using the intermediate start table 50 will be described.

When the image data is cut out, the data is read from the code data 40 and the intermediate start table 50 and executed. Here, the decompression processing is performed from the divided area closest to the desired cutout area. Therefore, first, the code start position (byte position, bit position) corresponding to the divided area to be decompressed is read from the intermediate start table 50. The read data is given to the binary image compression / expansion processing device 10 via the system bus 28. In the binary image compression / expansion processing device 10, the I / O instruction by the CPU 2 causes the byte position indicating the code start position and the bit position in the same byte to be I / O.
Predetermined register byte counter 2 in O register 18
0, set to pointer 24.

When the decompression process is activated and the code data 40 is input to the decoding circuit 14, the byte counter 20 is decremented for each 1-byte input. The input code data is skipped while the value of the byte counter 20 becomes "0". That is, decompression processing is not performed for unnecessary code data before the divided area including the image to be cut out. When the code data is sequentially input and the value of the byte counter 20 becomes “0”, it means that the byte including the code data at the head of the area where the desired decompression is to be started is fetched. Then, the pointer 24 indicates the decoding start position within the same byte.

If the system side can transfer the code data from the byte including the intermediate start dot position, the dot position can be set in the pointer 24 and the expansion can be immediately started from the sent byte.

First, when decompressing the leading line of the partial image, the horizontal mode instruction circuit 17 instructs the decompression processing in the horizontal mode. Since the leading line of each partial image is encoded with the code word in the horizontal mode when the code data 40 is generated, it can be expanded regardless of the pattern of the reference line. The decoding circuit 14 performs decoding from the decoding start position indicated by the content of the pointer 24. The generation circuit 16 generates an image according to the decoding result of the decoding circuit 14.

When decompressing the next line, the image data of the first line is used as a reference line and stored in the reference line image memory 27. The change point detection circuit 12 is
Reference line change point (b1 point) corresponding to the encoded line
Is detected from the image data of the reference line. on the other hand,
The decoding circuit 14 decodes from the bit position indicated by the content of the pointer 24. The generation circuit 16 generates an image according to the change point detection result of the change point detection circuit 12 and the decoding result of the decoding circuit 14.

Thereafter, images are sequentially generated until the number of lines of the input code data reaches a predetermined number. The decompression processing here is performed only for the partial image including the partial image to be cut out. Further, if the decoding of the required number of lines is completed in the middle of one partial image, the decompression process may be terminated there.

Next, the image data editing process in this system will be described. In this process, the portion to be edited is cut out from the binary image, and the change (deletion, addition, etc.) or replacement with another image is performed to obtain the code data 40 corresponding to the entire binary image again. It is what

FIG. 8 shows an outline of the procedure for editing a partial image. Partial images 1 and 2 are the partial images including the region to be cut out in the image data before editing in FIG. In order to expand the partial images 1 and 2, the number of lines of this partial image is set in the I / O register 18. Then, based on the code data (FIG. 8B) corresponding to the image data, expansion is performed by referring to the intermediate start / start table 50, and the expansion is terminated when the line counter 22 reaches a predetermined value.
As a result, only the necessary partial images 1 and 2 are cut out. This image data is stored in the main memory 4 and provided for editing processing. Area to be edited in the image thus partially cut out (FIG. 8C)
Is edited, for example, as shown in FIG.

Next, a method of obtaining code data including the image (area) edited in this way will be described.
First, referring to the intermediate start table value corresponding to the original image, the byte position and bit position of the partial image 1 are
It is set in the I / O register 18 and the pointer 26. Then, the code data up to a predetermined number of bytes, that is, the code data other than the partial images 1 and 2 to be edited is input and directly output to the outside. When the predetermined number of bytes (the head position of the partial image 1) is reached, the byte including the head of the partial image 1 is held in the register 16e of FIG. 4, and the head bit position is held in the pointer 26. In this state, the number of lines of the partial images 1 and 2 (range to be compressed) is set in the I / O register 18. Then, the image data of the edited partial images 1 and 2 is input, and compression processing is performed until the line counter 22 reaches a set value. However, when the compression process is performed on the first line, the horizontal mode is instructed by the horizontal mode instructing circuit 17, and the encoding in the horizontal mode is performed. The code data is connected at the position indicated by the pointer 26 in the generation circuit 16 and is generated as data to be output. When the code data up to a predetermined line is generated, the pointer 26 is in a state of indicating the bit position to be connected next.

For the partial images 3, ..., it is sufficient to connect the original codes before editing, but the code length may be changed by editing the images, and the bit positions to be connected may differ. Therefore, the code start position corresponding to the unedited image data 30 is read from the intermediate start table 50 and set in the pointer 24. Then, the code data is input from the byte including the head of the code data corresponding to the partial image 3 (when the code cannot be input from the middle of the code data 40 due to the condition of the system side, the code start position is set to the I / O register. Set it to 18 and skip unnecessary code data to get the required bytes). The decoding circuit 14 uses the funnel shifter 14b to read one byte (SHT) from the bit position indicated by the pointer 24.
D) is taken out and output to the generation circuit 16. Generation circuit 16
Then, the selector 16b causes the code data (SHT
D) is selected and the code data is connected at the position indicated by the pointer 26.

Further, since the code length may change when the partial image to be edited is coded, the code data is connected and the data indicating the code start position in the intermediate start table 50 is updated. To do. As a result, the code data 40 and the intermediate start table 50 corresponding thereto are obtained.

In this way, when the code data 40 corresponding to the image data 30 is generated, the head line of each partial image is coded by the code word in the horizontal mode, and the code start position (byte position) of each divided area is coded. ，
By storing (bit position) as an intermediate start table, the following effects can be obtained. That is, when the decompression process is performed based on the code data 40, since the leading line of each partial image 30-1, ... Is coded in the horizontal mode, it can be obtained as reference line data, and separately added to each partial image. There is no need to store the corresponding reference line data. Moreover, since the expansion processing can be performed for each partial image, a bitmap memory for storing the entire image is not required, and the processing time can be further shortened. Even when an image is edited, the compression process is performed only on the partial image including the edited area to generate the code data 40, so that the processing time can be shortened.

[0051]

As described above, according to the present invention, the code data in which the leading line of each partial image is encoded by a code word having no vertical correlation and the code of the code data corresponding to each partial image By storing the start position, even when a partial image is cut out from a large size image memory, it can be performed at high speed with a small memory capacity. Further, even when a partial image is edited, the decompressing process and the compressing process are performed only on the necessary image portion, so that high-speed processing is possible.

[Brief description of drawings]

FIG. 1 is a block diagram showing a detailed configuration of a binary image compression / expansion processing device 10 according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a system to which a binary image data cutout method according to the present invention is applied.

FIG. 3 is a block diagram showing a detailed configuration of a decoding circuit 14.

FIG. 4 is a block diagram showing a detailed configuration of a generation circuit 16.

FIG. 5 is a conceptual diagram of image data 30 corresponding to an original image.

6 is a conceptual diagram of two-dimensionally arranged code data 40 corresponding to the image data 30 shown in FIG.

FIG. 7 is a conceptual diagram of an intermediate start table 50.

FIG. 8 is a diagram for explaining an outline of a procedure for editing a partial image.

[Explanation of symbols]

2 ... CPU, 4 ... Main memory, 8 ... Mass storage device, 10
... binary image compression / expansion processing device 12, change point detection circuit,
14 ... Decoding circuit, 16 ... Generation circuit, 17 ... Horizontal mode instruction circuit, 18 ... I / O register group, 20 ... Byte counter, 22 ... Line counter, 24, 26 ... Pointer, 2
7 ... Reference line image memory, 28 ... System bus.

Claims (2)

[Claims] 1. A system for performing compression / expansion of a binary image, wherein a start line is not vertically correlated with a code word for each of a plurality of regions obtained by dividing the binary image by a predetermined number of lines. Encoding means for encoding other lines with a predetermined code word, first storage means for storing the code word obtained by the encoding means, and the first storage means for storing the code word. Second storage means for storing the code position data indicating the start position of the code word corresponding to each start line of the area in the code word, and the code position data stored in the second storage means, Binary image data, comprising: an image cutting-out unit that cuts out a partial image in the binary image for each of the regions based on the codeword stored in the first storage unit. Cutting method 2. A system for performing compression / expansion and editing of a binary image, comprising: first encoding means for encoding each of a plurality of regions obtained by dividing the binary image by a predetermined number of lines. Codeword storage means for storing the codeword obtained by the first encoding means, and start of the codeword corresponding to each start line of the area in the codeword stored by the codeword storage means Position storage means for storing code position data indicating a position, code position data stored in the position storage means, and in the binary image based on the code word stored in the code word storage means Image cutting means for cutting out a partial image for each area; second coding means for coding the binary image cut out by the image cutting means and edited for each area; By the second encoding means A code word combining means for connecting the code word thus obtained with a code word corresponding to a region not cut out by the image cutting means and storing the code word in the code word storing means. Binary image data cutout method.