JP2806169B2 - Compressed data decompression device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image compression apparatus, for example, M
The present invention relates to a compressed data decompression device for decompressing image data compressed by an H-type image compression device.

[0002]

2. Description of the Related Art Modified Hu (MH) is a standard coding method for group 3 (G3) facsimile.
ffman) method. This is called a one-dimensional encoding method, and is a method in which image data is converted into continuous data of codes having different bit lengths and compressed. Generally, the continuation of the same signal is called a run. The image data before compression is data in which white or black runs in which white pixels and black pixels are continuous alternately exist, and the compressed data is the number of continuous white or black runs. Is continuous data of a code obtained by converting the run length as described above. The maximum number of bits of each code is 13 bits, and an EOL (End Of Line) code is added after the data of one horizontal scanning line.
It is promised that the EOL code is added six times at the end of the compressed data.

FIG. 4 is a flowchart showing a conventional compressed data decompression apparatus for decompressing image data compressed by the MH system. Note that “add” described in the figure is a position at which compressed data is extracted. Next, the operation will be described. When the data decompression process starts (step 50), the compressed image data is continuous data in which codes having a maximum length of 13 bits are continuous. Thirteen bits are temporarily extracted (step 51), and a table prepared in advance in which run lengths and codes correspond one-to-one is read (step 52). Since the extracted compressed data does not always include a valid code, that is, a code corresponding to a run-length or EOL code, the above table is compared with the compressed data to determine a code corresponding to the table. Is determined (step 53), and if the corresponding code exists on the table, the code is set to EO.
It is determined whether it is an L code (step 54). If it is not the EOL code, it is not the end of the horizontal scanning line.
The extraction position add of the compressed data is updated (step 55). Data extraction is performed in step 51 at step 13
Since the bits are temporarily extracted, the original code length of the code existing in the extracted compressed data is determined in step 5.
In the collation of No. 3, the data is read from the table, and the compressed data extraction position add is updated by the original code length, thereby updating to the next correct extraction position.

[0004] Thus, the decompression process for returning one code in the compressed image data to the run length is completed, and the uncompressed image data is restored from the read white or black run length and stored in the memory ( Step 56). In this manner, the decompression and restoration processing of one code is completed, and is still in the middle of the horizontal scanning line. Therefore, the process returns to the extraction of 13 bits again for the decompression processing of the next code (step 5).
1) The above processing is repeated.

If the code is an EOL code in step 54, it is determined whether the current next code is also an EOL code (step 57). If the next code is not the EOL code, it means the end of the horizontal scanning line, so the compressed data extraction position add is shifted by the code length of the EOL code and updated to the beginning of the next horizontal scanning line (step 5).
8) Returning to the extraction of 13 bits, the decompression process is performed again (step 51). If the next code is also an EOL code in step 57, the end of the image data is determined.
End processing is performed (step 59). Step 53
Also, if the corresponding code corresponding to the compressed data extracted for some reason does not exist in the table, the termination processing is performed (step 59).

[0006]

In such a conventional compressed data decompression apparatus, when a part of the compressed data is broken due to a defective writing of the compressed data or a defective reading of the compressed data, a step 53 in FIG. In the processing, since it is determined that the corresponding code does not exist on the table and the end processing is performed, there is a problem that data below the defective data cannot be read even if it is normal data. An object of the present invention is to provide a compressed data decompression device that can read normal data even if compressed data including defective data is extracted.

[0007]

SUMMARY OF THE INVENTION The present invention is directed to an MR system.
A data extraction unit that extracts compressed data from the image data that has been encoded according to the above is compared with a table that is a code table and the compressed data that is output from the data extraction unit, and the MH code or MH code included in the compressed data is compared. and it outputs the data decompression by reading the data corresponding to the MR code from the table, if the code corresponding to the table does not exist and the collation unit for outputting an instruction signal shifts to bad data processing, the next line Indicates that MH encoding was performed
First tag bit, next line is MR encoded
Of the second tag bit indicating the
The terminal code indicating the terminal and the data output from the collating unit
Are equal or not, and based on the result, the matching unit
The code length, which is the number of bits of the code corresponding to the data from
Output and collate if the collation unit shifts to bad data processing
Terminal code determination unit that repeatedly outputs an addition signal until the data from the unit becomes equal to the first tag bit, and stores the instruction signal output from the collation unit, the code length output from the terminal code determination unit, or the addition signal. The data extracting unit is characterized in that the data extracting unit has an extracting position calculating unit that updates the extracting position of the compressed data and specifies the extracting position of the compressed data in the compressed image data.

[0008]

According to the present invention, when defective data is included in one part of the compressed data, the defective data portion can be skipped by repeatedly updating the extraction position of the compressed data until the end code is reached. it can.

[0009]

FIG. 1 is a block diagram of a compressed data decompression device showing an embodiment of the present invention, and FIG. 2 is a flowchart showing the same. In FIG. 1, reference numeral 1 denotes a data extracting unit for temporarily extracting 13 bits, which is the maximum length of a code, from the compressed image data a, and 2 denotes a table in which run lengths and codes are described in a one-to-one correspondence. This is a table storage unit for storing. Reference numeral 3 denotes a collation unit, and a table storage unit 2
Is compared with the compressed data c output from the data extracting unit 1, and if a corresponding code exists on the table, a run length d1 corresponding to the code is output. If there is no corresponding code, an instruction signal f for adding one bit to the extraction position add of the compressed data is output. Reference numeral 4 denotes a terminal code determination unit, and the run length d1 output from the collation unit 3 is EOL.
A code length d2, E for updating the compressed data extraction position add based on the result of the determination.
The code length e1 of the OL code or the addition signal e2 for adding one bit is output, or the run length d1 is output as it is. Reference numeral 5 denotes an extraction position calculation unit which extracts the compressed data according to the instruction signal f output from the collation unit 3, the code length d2 output from the end code determination unit 4, the code length e1 of the EOL code, or the addition signal e2. add
And outputs to the data fetching unit 1 a fetched position add indicating the fetching of the compressed data and its position. Reference numeral 6 denotes a data end determining unit for extracting the code next to the code currently being processed based on the code length e1 of the EOL code output from the terminal code determining unit 4 and determining whether to end the image data.

Next, the operation of the compressed data decompression device of the present invention will be described with reference to the flowchart shown in FIG. The data extracting unit 1 temporarily extracts 13 bits from the extracting position in the compressed image data a input from an external device (not shown), that is, the position indicated by the extracting position add of the compressed data stored in the extracting position calculating unit 5 ( FIG. 2 step 11). The compressed data c output from the data extracting unit 1 is collated by the collating unit 3 with the table b stored in the table storage unit 2, and if a corresponding code exists on the table, the run corresponding to the code is performed. The code is converted to length d1 and output from the collating unit 3 together with the original code length d2 of this code present in the compressed data (steps 12 and 13). The end code determination section 4 determines whether or not the run length d1 is equal to the EOL code. If the run length d1 is not the EOL code, the end length determination section 4 outputs the run length d1 to an image data restoring device (not shown) and outputs the original code length d.
2 is output to the extraction position calculation unit 5 (step 1).
4). The extraction position calculation unit 5 calculates the original code length d2
The compressed data extracting position add is shifted and updated by an amount corresponding to the extracted compressed data, and the extracting position add that instructs the data extracting unit 1 to extract the next compressed data and its position is output (step 15).

[0011] Thus, the decompression process for returning one code in the compressed image data a to the run length d1 is completed, and the image data restoring device outputs the uncompressed image from the input white or black run length d1. The data is restored (step 16). In this way, one code is expanded,
Since the restoration process is completed and it is still in the middle of the horizontal scanning line,
In response to an instruction from the extraction position calculation unit 5 to the data extraction unit 1 (step 15), the process returns to the extraction of 13 bits again (step 11), and the above processing is repeated.

At the end of the horizontal scanning line or the end of the image data, the extracted compressed data is an EOL code, so the end code determination unit 4 sends the EOL code to the data end determination unit 6.
The code length e1 of the code is output (step 14). The data end determination unit 6 always receives the current compressed data extraction position add from the extraction position calculation unit 5 and stores it internally as the extraction position add2. However, when the code length e1 of the EOL code is input, the data end determination unit 6 Is updated by shifting the extraction position add2 stored in the image data a by the above-described code length e1, and extracting 13 bits from the position of the compressed image data a indicated by the extraction position add2, that is, extracting the bit including the next code to the EOL code. It is determined whether it is (step 17).

If it is not an EOL code, it can be determined that the end of the horizontal scanning line has been reached.
Outputs the code length e1 of the EOL code to the extraction position calculation unit 5, and shifts the extraction position add of the compressed data by the code length e1 to update to the beginning of the next horizontal scanning line. The fetch position calculation unit 5 outputs the fetch position add instructing data fetch to the data fetch unit 1 (step 18), and causes the data fetch unit 1 to fetch 13 bits (step 11). If it is the EOL code in step 17, it can be determined that the image data has ended.
The data end determination unit 6 performs end processing (step 1).
9).

The operation up to this point is the same as the operation of the compressed data decompression device shown in FIG.
3, it is determined that the corresponding code does not exist on the table, so that the operation shifts to the failure mode operation, the failure mode is maintained, the failure mode signal d3 is output to the terminal code determination unit 4, and the compression is performed to the extraction position calculation unit 5. An instruction signal f for instructing to add one bit to the data extraction position add is output.

The extraction position calculation unit 5 updates the extraction position add of the compressed data by adding one bit, and outputs the extraction position add indicating the extraction from the compressed image data a and the position to the data extraction unit 1. To extract 13 bits. (Step 20).
In the case of the failure mode operation, if the code corresponding to the compressed data c output from the data extracting unit 1 is on the table, the matching unit 3 outputs the run length d1 corresponding to the code to the terminal code determination unit 4, If it does not exist, it outputs appropriate data that has no meaning set in advance. The termination code determination unit 4 receives the failure mode signal d3 from the verification unit 3 and shifts to the failure mode, and determines whether or not the run length d1 output from the verification unit 3 is equal to the EOL code (step 21). If it is not the EOL code, an addition signal e2 of 1-bit addition is output to the extraction position calculation unit 5.
The extraction position calculation unit 5 instructed to add 1 bit updates the extraction position add of the compressed data again by shifting by 1 bit, and instructs the data extraction unit 1 to extract 13 bits from the compressed image data a ( Step 2
0).

The operation of such a failure mode is repeated until the end of the EOL code is reached. When the end of the EOL code is reached, the terminal code determination unit 4 releases its own failure mode and then sends a failure mode release signal g to the verification unit 3. To output the code length e1 of the EOL code to the extraction position calculation unit 5 (step 21). The fetch position calculation unit 5 updates the fetch position add of the compressed data by the code length e1 of the EOL code to the beginning of the next horizontal scanning line, and causes the data fetch unit 1 to fetch the compressed image data from the image data. Instructed take-out position add
Is output (step 18), 13 bits are taken out, and the operation is returned to the normal data decompression operation (step 11). That is, when the defective data is taken out in the middle of the horizontal scanning line, the collating unit 3 and the end code determining unit 4
Shifts to the failure mode operation, skips the failure data portion by repeating the one-bit update of the extraction position of the compression data until the EOL code is reached, and resumes the expansion of the compressed data from the horizontal scanning line having normal data can do.

Although the example shown in FIG. 1 is applied to the MH system, the present invention can also be applied to an MR (Modified Read) system. In the MR system, only one line out of a certain K lines is subjected to MH encoding (hereinafter referred to as MH encoding), and the remaining K-1 lines encode differences from the previous line. (Hereinafter referred to as MR encoding) to compress image data. An EOL code is added to the end of each coded line, followed by a tag bit of "1" or "0" indicating whether the next line was MH coded or MR coded, respectively. . Hereinafter, the respective codes are EOL + 1 code, EOL
+0 sign. EOL + 1 at the beginning of the image data
It is promised to add a code, and to add the EOL + 1 code six times at the end.

FIG. 3 is a flow chart of a compressed data decompressing apparatus showing another embodiment of the present invention for decompressing image data compressed by the MR method. The compressed data decompression device of the present invention has basically the same configuration as the example of FIG. 1, and outputs run lengths to different image data decompression devices for MH-encoded lines and MR-encoded lines. MH encoding or M
The only difference is that it is performed by the image data decompression device for R encoding.
The only difference is that it corresponds to the R method.
2 is the same as FIG.

If the corresponding code exists in the table in step 33, it is determined whether or not the run length read from the table is the EOL + 1 code (step 34). EOL + 1 is added to the head of the compressed image data.
Since the code has been inserted, if it is the head of the compressed image data, the process proceeds to the determination of whether the next code is the EOL + 1 code (step 36). Since it is not the end of the image data, the next code is not the EOL + 1 code, and thus can be determined as the MH coded line.
The process is shifted to an MH encoding image data restoration device (not shown). In the MH coding image data decompression device, a command is issued to the same data retrieving unit as in the example of FIG. 1 to retrieve one line of compressed data, decompress the image data, and set the compressed data retrieval position add for one line. , And updates the code length up to the EOL code of the next line (step 37).

Thus, the extension processing of the MH-encoded line is completed, and the next horizontal scanning line is taken out (step 31). Since the next horizontal scanning line is a line that has been MR-encoded, it can be determined in step 34 that it is not an EOL + 1 code, and processing is transferred to an MR encoding image data restoration device (not shown) to perform processing for one line in the same manner as described above. (Step 35).

If defective data occurs in the EOL code portion, it is determined in step 33 that the corresponding code does not exist on the table, and the compressed data extracted by the same operation as in the example of FIG. Until the compressed data extraction position is updated and repeated (step 4
0, 41). That is, in the MR method, the difference from the previous line is encoded.
The encoding line becomes meaningless data. Therefore, when the defective data is extracted, the extraction position add of the compressed data is shifted until the EOL + 1 code at the head of the MH encoding line is confirmed, and the process returns to the normal processing when the EOL + 1 code is confirmed. The normal data portion can be expanded and restored by skipping the portion. Although not shown in the flowchart of FIG. 3, if a loop similar to steps 40 and 41 is provided in steps 35 and 37, the occurrence of defective data in the middle of the line can be skipped. .

[0022]

According to the present invention, even if defective data is included in one part of the compressed data, the defective data part is skipped by repeatedly updating the extraction position of the compressed data until the end code is reached. Therefore, if there is normal data after the defective data, the normal data can be extracted and expanded.

[Brief description of the drawings]

FIG. 1 is a block diagram of a compressed data decompression device showing one embodiment of the present invention.

FIG. 2 is a view showing a flowchart of the compressed data decompression apparatus of FIG. 1;

FIG. 3 is a diagram showing a flowchart of a compressed data decompression device showing another embodiment of the present invention.

FIG. 4 is a diagram showing a flowchart in a conventional compressed data decompression device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Data extraction part 2 Table storage part 3 Collation part 4 Termination code judgment part 5 Extraction position calculation part 6 Data end judgment part a Image data after compression add Extraction position of compressed data b Table c Compressed data d1 Run length d2 Code length d3 Failure mode signal e1 Code length of EOL code e2 Addition signal f Instruction signal g Failure mode release signal

Claims (1)

(57) [Claims] A data extraction unit for extracting compressed data from compressed image data encoded by an MR method , a table as a code table, and compressed data output from the data extraction unit, and The data corresponding to the MH code or the MR code included in the data is read out from the table and the data is decompressed and output. If the corresponding code does not exist on the table, the process proceeds to the defective data processing and the instruction signal is output. A collating unit to be output, and a first tag video indicating that the next line has been MH encoded.
, A second indicating that the next line is MR encoded
Tag bit or end indicating end of compressed image data
Whether the end code is equal to the data output from the collating unit
Data from the collation unit based on the result.
Outputs the code length, which is the number of bits of the code corresponding to
If the part shifts to bad data processing, the data
A termination code determination unit that repeatedly outputs an addition signal until the data becomes equal to the first tag bit , an instruction signal output from the collation unit, and a code length or an addition signal output from the termination code determination unit. A compressed data decompression device, wherein the data retrieval unit updates a retrieval position of the compressed data and specifies a retrieval position of the compressed data in the compressed image data.