JPS6333350B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an image storage device for characters, patterns, etc., and particularly to a storage device for encoded image data.

Conventionally, most image data handled in the field of image processing has been digital data because of ease of processing. Since the data has a two-dimensional structure and contains density information, the number of bits that make up the data is quite large. In addition, the image has a scale,
When higher resolution and gradation are required, the number of bits increases dramatically. Therefore, storing such image data requires a storage device with a large capacity. There are currently existing magnetic disks,
Although magnetic tapes and the like are used, if the number of stored images is increased and the access time is shortened, it is necessary to consider increasing the information density of image data. Therefore, it is conceivable to utilize data compression, which is currently used in image transmission devices, particularly in the field of facsimile, to reduce the amount of image data transmitted, to improve information density.

Several encoding methods have been developed for the data compression mentioned here, but among them, the encoding method that correlates multiple lines is particularly effective in terms of compression rate for images with monotonous patterns. and
By using this method, it is possible to store image data with high information density. However, this method is not an encoding method that completes line by line, so when you want to perform processing such as cutting out parts of an image that has been encoded, you have to decode a large amount of image data that is not directly related. Although the information density has increased, the processing efficiency has become extremely low.

It is an object of the present invention to eliminate the above-mentioned drawbacks, to provide an encoding function and a storage function that allow for short access time and efficient partial extraction of image data despite having a high information density. An object of the present invention is to provide a coded image storage device having the following features.

According to the present invention, image input means binarizes image density data obtained by scanning a two-dimensional image and outputs the binarized data serially line by line, and an encoding method is selected for the image data obtained by the image input device. An encoding device that selects and encodes either an encoding method that completes in one line or an encoding method that correlates to multiple lines depending on the signal, and image data obtained by the encoding device, The first line of each segment is divided into segments of n lines in order, and the first line of each segment is encoded using a coding method that completes with one line, and the remaining lines are encoded using a code that correlates with multiple lines in each segment. an encoding control device that outputs an encoding method selection signal to the encoding device so that the encoding method can be encoded; and a device that stores part or all of the compressed image data obtained from the encoding device in a buffer memory. , Storage control that stores the start position of line data encoded using an encoding method that completes in one line, that is, the start position of each segment, in the buffer memory in segment order on the stored compressed image data, and creates an address table. When decoding compressed image data, which line is coded to be completed in one line? It is possible to create a data structure that can be decrypted even from the beginning.

Next, the present invention will be explained with reference to the drawings. Hereinafter, the numbers attached to signal lines in the drawings will be explained as referring to the signals on the signal lines.

Figure 1 shows a block diagram. The image input device 1
An image scanner such as FSS or ITV can be used, and if the scanned image data is multivalued,
A circuit for binarizing multivalued data is provided at the location of the image scanner. The encoding device 2 converts the serial image data from the image input device 1 into an encoding method that completes in one line (for example, a run-length encoding method) and a plurality of lines according to the encoding method selection signal from the encoding control circuit 5. This is an apparatus that performs encoding by selecting one of the encoding methods (for example, sequential encoding method) that have a correlation with the data. The encoding control device 5 divides the image data output from the image input device 1 into segments each having n lines based on a pre-specified number n, and the first line of each segment is completed with one line. This is a device that generates an encoding method selection signal that specifies an encoding method and can designate, for the remaining lines, an encoding method that is correlated to a plurality of lines on a segment-by-segment basis. When the storage control device 3 stores part or all of the encoded image data from the encoding device 2 in the buffer memory within the storage control device 3, the storage control device 3 detects the encoding method using the method described later and performs the following on the encoded image data. This is a device that stores the start position of the encoded data that is completed in one line, that is, the start position of the segment, and creates an address table.The storage device 4 is a device that stores the encoded image data and the address table. .

Before entering into a detailed block description of the present invention, the procedure and data structure by which image data is encoded and stored according to the present invention will be explained based on an embodiment. Figure 2 shows the data structure. FIG. 2a is a conceptual diagram of the original image data 9 that is output from the image input device 1 and is displayed in correspondence with the original image, and FIG.
Fig. 3c shows a conceptual diagram of the compressed image data 14 arranged two-dimensionally in the memory of , and c shows a conceptual diagram of the address table 21 in the memory of the storage control device 3.

When encoding the original image data 9, instead of encoding the entire image data all at once, it is divided into segments 10 and 11 of n lines each based on a pre-specified number n, and encoded in segment units. We will adopt a method of converting

The original image data 9 extracted in segments is encoded using the following two types of encoding methods depending on the lines within the segment. First, the first lines 12 and 13 of each segment are encoded using a run-length encoding method, and the remaining lines of the segment are encoded using a sequential encoding method. The sequential encoding method mentioned here basically involves checking the connection state between the black run signal of the encoded line, which is the current scan line of the image, and the black run signal of the reference line, which is the previous scan line.
This is a method of encoding it. Examples of this encoding method include Japanese Patent Application No. 52-82247 (Japanese Patent Application Laid-open No.
No. 17615) There is an image encoding device in the specification.

The data encoded in this way becomes as shown in FIG. 2b. Generally, the length of encoded data changes depending on the data content even if data of the same length is encoded.
7,19 or run-length encoded line 1
There is no regularity in the first addresses 17 and 19 of 8 and 20. Therefore, during encoding, the start addresses 17 and 19 of each segment are stored in the address table 21 and the segment start address storage devices 22 and 23, and the segments 10 and 19 of the original image data 9 are
11 and segment 15 of compressed image data 14,
16 in one-to-one correspondence. Therefore, when cutting out a partial image from the compressed image data 14, if the segment containing the first line of the partial image to be cut out is known in the original image data 9, the starting addresses 17 and 19 of the corresponding segment in the compressed image data 14 are determined. By referring to the address table 21, decoding is started from that address, and when decoding of the required number of lines is completed, decoding is stopped there, allowing efficient extraction of the necessary data.

Next, one embodiment of the present invention will be described.

FIG. 3 shows a block diagram of the encoding device 2. The image scanned by the image input device 1 is serially input line by line to both the run length encoding circuit 30 and the sequential encoding circuit 31 in the encoding device 2 as binary original image data 9. . Each encoding circuit 3
0 and 31 operate simultaneously in synchronization with the input clock signal 102 from the image input device 1 to encode the original image data 9. Since each encoding circuit has a different encoding speed, its output (encoded image data 3
01, 311) are once taken into the respective buffers 32, 33 in synchronization with the encoding circuit output clock signals 302, 312, and both buffers are 1
After the encoded image data for the line is stored, control is transferred to the next stage selection circuit 34.

Here, according to the block diagram of FIG.
The details of 2 and 33 will be explained by taking the run-length encoding circuit buffer 32 as an example. The run-length encoding buffer 32 consists of a so-called buffer memory (shift register 35) and other control sections.

The run-length encoded image data 301, which is the output of the run-length encoder 30, is transferred to the shift register 3 in synchronization with the run-length encoder output clock signal 302 (via the OR gate 39).
It will be incorporated into 5. The shift register 35 takes in encoded image data line by line as described above, and its control is performed by detecting a line separation signal from the serially continuous encoded image data.

Specifically, in the encoded image data, for example,
If it is decided in advance at the time of encoding that k consecutive zeros are a line delimiter code, the lower k bits of the shift register 35 (k bits on the input side) can be transferred to k inverters (41-1, 41-1).
2, 41-3,..., 41-k) and AND gate 4
2 can be used to detect a line break (line synchronization signal 421), and the line synchronization signal 4
21 can be used to output an operation stop signal 325 to the run-length encoding circuit 30 and store one line of encoded image data in the shift register 35. However, since the length of one line of encoded image data varies depending on the content of the original image data, empty bits of various lengths are generated on the output side of the shift register 35. In order not to output unnecessary bits during output, these empty bits must be shifted. To do this, the flip-flop 37 is set by the line synchronization signal 421 to operate the oscillator 38, and the clock signal 381 is used to shift the contents of the shift register 35. Here, the number of shifts is determined by counting the number of encoded image data, that is, the number of shifts at the time of data input, in advance in the counter 36, and using the complement of the number.

The signals exchanged between the main buffer 32 and the next stage selection circuit 34 include encoded image data 3
21, a readout clock 341, the number of encoded image data 322, a strobe signal 323, and a line storage end signal 324, and a scanning invalidation signal 326 between it and the preceding image input device 1. This scanning invalidation signal 326 is a signal that prevents the image input device 1 from advancing scanning beyond the current scanning line and prohibits data output.

Run-length encoding circuit buffer 3 described above
The operation of 2 is the same as that of the sequential encoding circuit buffer 3.
This also applies to 3.

The operation of the selection circuit 34 after one line of encoded image data is stored in the two encoding circuit buffers 32 and 33 will be described with reference to FIG.
The two types of line storage completion signals 324 and 334 from the encoding circuit buffers 32 and 33 set flip-flops 47 and 48, respectively, and the output signals and the encoding method selection signal 521 from the encoding control device 5 are sent to flip-flops 47 and 48, respectively.
AND gate 42 is taken and the selected one (now,
This is referred to as a run-length encoding method).

Its output, the read clock signal 341, is
It is sent to the encoding circuit buffer 32 and requests the encoded image data 321 for one line of the selected one. At this time, the end of importing the encoded image data is 32, which is the number of data sent from the encoding device 2 in advance.
2 is performed by counting the readout clock 341 with the counter 44 latched by the strobe signal 323, outputting a carry down signal 441, resetting the flip-flop 47, and stopping the oscillator 43 from oscillating.

The above-mentioned encoding method selection signal 521 is obtained by ORing (OR gate 49) the two types of line storage end signals 324 and 334 from the encoding circuit buffers 32 and 33 (detecting the later end of storage). This is a signal generated by the encoding control device 5 based on .

Encoded image data 34 output from the main selection circuit 34
3 is a system in which the read clock is sent to the selected one of the run-length encoding circuit buffer 32 and the sequential encoding circuit buffer 33, and only the one that requests it outputs the encoded image data. Output 3 of conversion circuit buffers 32 and 33
It's just ORing 21,331. The same goes for the selection circuit output clock 344.

Next, the encoding control device 5 will be explained using the block diagram of FIG.

This device has two registers 51 and 53 and one program counter 52, and an encoding control circuit 5.
and the encoding method selection signal 52 as described above.
1 to the encoding device 2. The program register 51 receives an image start signal 103 from the image input device 1, and uses it as a strobe signal for a prespecified encoding parameter (n: number of lines in a segment) 111.
Incorporate. Each time the n-ary counter 52 programmed thereby counts n line storage end signals 346 from the encoding device 2, that is, every n lines of original image data, it outputs a carry signal, and uses it in the encoding method. A selection signal 521 is used. Generally, the carry signal is output when the n-ary counter finishes counting n clocks, so in order to output it at the beginning of the n line storage end signal as in this circuit, the register 53 must be set in advance to the parameter. (n-1) 112 is set,
The image start signal 103 from the image input device 1
This is done by initializing the n-ary counter with .

Next, the storage control device 3 will be explained.

FIG. 7 shows a block diagram of one embodiment. Encoded image data 343 from the encoding device 2 is written into the buffer memory 61 in response to a clock signal 344. At the same time, two shift registers 55 and 56 are used to create an address table.
(for mode bit detection and data end signal, respectively) are also input. The mode bit detection shift register 55 is used to determine whether encoded image data is run-length encoded data or sequentially encoded data.
If the two bits following a line separator code with k consecutive zeros are 10, the following data is run-length encoded data, and if it is 11, the following data is sequentially encoded data. If you install k+1 inverters (57-0, 57-1,
…57-k) and AND gates 58 and 59 to detect the start of run-length encoded data 591
can be created. At the same time as this signal is output, the contents of the buffer memory address counter 62 are stored in the address table buffer 63, so that the start address of the run-length encoded data on the encoded image data, that is, the address where the start addresses of the segments are sequentially stored. The table is completed. The encoded image data in the buffer memory 61 and the address table in the address table buffer 63 created as described above are sequentially transferred to the storage device 4 upon completion of the data. The data end signal 601 for this purpose is 1 for encoded image data.
If it is determined in advance at the time of encoding that the data is complete if l consecutive numbers of the data are consecutive, the shift register 56 for data termination and the AND gate 60 can be used.

When the storage device 4 receives the data end signal 601, it outputs an address table request clock 702, takes in the address table 631, then outputs an encoded image data request clock 701, and takes in the encoded image data 611.

FIG. 8 shows m-th and m+1-th address tables 801 and 803 and m-th and m+1-th encoded image data 802 and 804 stored in the storage device 4 using the magnetic tape 80 as an example.

In addition, the above-mentioned storage control device 3 can also use a computer as another embodiment. This can be achieved by associating the buffer memory with the main memory of the computer and the other parts with programs or interfaces. Further, the storage device 4 can be any standard auxiliary storage device (magnetic tape, magnetic disk, etc.) for computers.

The capacity of the buffer memory 61 in the storage control device 3 described above is optimally large enough to accommodate one image, but even if it is small, the data end signal 61
A similar operation can be performed by simply replacing 1 with a signal that detects when the memory becomes full and adding a few control sections.

As described above, the encoded image storage device according to the present invention has the function of adding information that allows partial extraction to encoded image data. It is possible to efficiently cut out partial images from the compressed data obtained by the invention.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of the present invention, FIGS. 2 a, b, and c are conceptual diagrams of the data structure showing the steps for encoding and storing image data when the present invention is used. The figure is a block diagram for explaining an encoding device 2 which constitutes a part of an embodiment of the present invention and which encodes image data. A block diagram for explaining an encoding circuit buffer which is a circuit buffer,
FIG. 5 is a block diagram for explaining a selection circuit forming a part of the encoding device 2 and selecting encoded image data in two types of encoding circuit buffers;
The figure forms part of an embodiment of the present invention, and is a block diagram for explaining the encoding control device 5 that controls the encoding device. , a block diagram for explaining a storage control device 3 that receives encoded image data from an encoding device, creates an address table, and stores the encoded image data and address table in a storage device. FIG. 3 is an explanatory diagram of an address table and a storage form of encoded image data when used as a magnetic tape. In the figure, 1...image input device, 2...encoding device, 3...storage control device, 4...storage device, 5...
Encoding control device, 6... Image, 7... Encoded image data, 8... Encoding parameter, 9... Original image data,
10...Original image data first segment, 11...Original image data second segment, 12...Original image data first segment first line, 13...Original image data second segment first line, 14...Encoded image data, 15 ...Encoded image data first segment, 1
6...Encoded image data second segment, 17...Encoded image data first segment start address, 1
8... Original image data first segment first line on encoded image data, 19... Encoded image data second segment start address, 20... Original image data second segment first line on encoded image data, 21...Address table, 22...First segment start address storage position, 23...Second segment start address storage position, 30...Run length encoding circuit, 31...Sequential encoding circuit, 32...Run length encoding circuit buffer, 33 ... Sequential encoding circuit buffer, 34... Selection circuit, 35... Shift register, 36... Counter, 37... Flip-flop, 38... Oscillator, 39, 40... OR gate,
41-1, 41-2,... 41-k... Inverter, 42... AND gate, 43... Oscillator, 44...
counter, 45... oscillator, 46... counter,
47, 48... flip-flop, 49... OR gate, 50... monostable multivibrator, 51...
Program register, 52...Programmable counter, 53...Register, 55, 56...Shift register, 57-0, 57-1,...57
-k...Inverter, 58, 59...AND gate,
60...AND gate, 61...Buffer memory for encoded image data, 62...Address counter, 6
3...Address table buffer, 64...Address register, 80...Magnetic table, 101...Original image data, 102...Original image data output clock, 103...Image start signal, 111...Encoding parameter n, 112...Encoding parameter n −
1, 301...Run length encoded data, 302
...Run length encoded data output clock, 31
1... Sequentially encoded data, 312... Sequentially encoded data output clock, 321... Run length encoded data, 322... Number of bits per line of run length encoded data, 323... 322 Strobe signal, 3
24...Run-length encoding data line storage end signal, 325...Run-length encoding stop signal, 3
26...Scanning invalid signal, 331...Sequentially encoded data, 332...Number of bits per line of sequentially encoded data, 333...332 strobe signal, 334...Sequentially encoded data storage end signal, 335...Sequentially encoded stop signal, 336...Scanning Invalid signal, 341... Run length encoded data read clock, 342
...Sequential encoded data read clock, 343...Encoded image data, 344...Encoded data output clock, 346...1 line encoded image data storage end signal, 361...Shift end signal, 381...Shift clock, 421...Line synchronization signal ,441,
461... Decrement signal, 511... Program data, 521... Encoding method selection signal, 591... Run length encoded data head detection signal, 601... Encoded image data, 631... Address table, 7
01...Encoded data request clock, 702...Address table request clock, 801...mth address table, 802...mth encoded image data, 803...m+1th address table,
804...m+1-th encoded image data are respectively shown.

Claims (1)

[Scope of Claims] 1. Image input means that binarizes image density data obtained by scanning a two-dimensional image and outputs the binarized data serially line by line, and an encoding system for the image data obtained by the image input means. an encoding means for compressing and encoding the image by selecting one of two encoding methods, an encoding method that completes in one line and an encoding method that correlates with a plurality of lines, according to a selection signal; The image data obtained by the input means is divided into segments of n lines sequentially from the first line, and the first line of each segment is encoded using an encoding method that is completed in one line.
encoding control means for outputting an encoding method selection signal to the encoding device so that the remaining lines can be encoded by an encoding method having a correlation with the plurality of lines in units of segments; In the compressed image data, part or all of the continuous data from the front end or the rear end is stored in the buffer memory, while the start position of the data encoded by the encoding method that is completed in one line on the compressed data is stored in the buffer memory. That is, it has a storage control means for storing the start position of each segment in a buffer memory in a one-to-one correspondence with that segment to create an address table, and a storage means for storing the compressed image data and the address table. A coded image storage device characterized by: 2. The encoding device includes a run-length encoding circuit, a sequential encoding circuit, and a signal selection circuit,
2. The encoded image storage device according to claim 1, wherein each input image data is encoded and one of the encoded signals is outputted according to a selection signal.