JPH05128101A - Data compression and restoration system - Google Patents

Abstract

PURPOSE:To effectively utilize the capacity of a dictionary as a whole by increasing the capacity of a split dictionary in accorcance with the number of registration at the time of encoding or deoding. CONSTITUTION:An initial set software 12 sets a minimum size to the respective sharing dictionaries 10-1 to 10-255 which are split up into 256, e.g. at the time of encoding or decoding and initial-registers all the character kinds of 256 by putting indexes in each character unit. A dictionary retrieving software 14 specifies a specific split dictionary 10-(i) with a history based on the group of last characters among character strings which are encoded just before then, for example, the character code of the last character at the time of encoding an input character string, and retrieves a partial string whose length coinsides with maximum length among the partial strings which are registered in a specified partial string 10-(i) and already encoded. A dictionary capacity increasing software 18 increases its dictionary capacity when the index of the split dictionary 10-(i) exceeds the prescribed size of a dictionary and keeps the size of the dictionary as it is when the index does not exceed it.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data compression and decompression method using an LZW code known as an improvement of an incremental decomposition type which is a kind of universal code. In recent years, various data such as character codes, vector information, and images have been handled by computers, and the amount of data handled has been increasing rapidly. When handling a large amount of data, omitting redundant parts of the data and compressing the amount of data reduces the storage capacity and enables faster transmission.

Universal encoding has been proposed as a method of compressing various data by one method.
Here, the field of the present invention is not limited to compression of character codes and can be applied to various data. However, in the following description, the word used in information theory is followed, and one word unit of data is called a character. , Let's call a character string a string of data connected by an arbitrary number of words.

As a typical method of the universal code,
There is Ziv-Lempel code (for details,
For example, see Munakata "Ziv-Lempel Data Compression Method", Information Processing, Vol.26, No.1, 1985). For Ziv-Lempel codes, two algorithms have been proposed: a universal type and an incremental decomposition type (Incremental parsing).

Further, as an improvement of the universal type algorithm, LZSS code, (TC Bell, "Better OPM / L
Text Compression ”, IEEE Trans. On Commun., Vol.C
OM-34, No. 12, Dec. 1986). Further, as an improvement of the incremental decomposition type algorithm, LZW (Lempel-Ziv-Welc
h) Signed (TA Welch, “A Technique for High-
Performance Data Compression ”, Computer, June 1984
reference).

Among these codes, the LZW code has come to be used for file compression of a storage device because of the high speed processing and the simplicity of the algorithm.

[0006]

2. Description of the Related Art A detailed algorithm for encoding / decoding a conventional LZW code is shown in FIG. 9, and a detailed algorithm for decoding is shown in FIG. The LZW encoding has a rewritable dictionary, divides the data consisting of input character codes into different character strings, assigns numbers to the character strings in the order in which they appear, registers them in the dictionary, and is currently inputting them. The character string is represented by the number of the longest matching character string registered in the dictionary and is encoded.

In the LZW encoding process of FIG. 9, first, in step S1, a character string consisting of one character for all characters is registered in advance as an initial value, and then encoding is started. In step S2, the input first character K is used as a reference number ω for dictionary search, and this is used as a prefix character string (prefixstring). Next, in step S3, the next character K of the input data is read, and in step S4, the character string (ωK) obtained by adding the character K read in step S3 to the initial character string ω obtained in step S2.
Is searched for in the current dictionary.

If the character string (ωK) is found in the dictionary in step S4, the character string (ωK) is replaced with the reference number ω in step S5, and it is determined in step S5 whether or not the input data has ended. Returning to this, the search for the maximum matching length is continued until the character string (ωK) cannot be searched from the dictionary.
Next, in step S4, if the character string (ωK) is not in the dictionary,
Proceeding to step S7, the reference number ω of the character K obtained in step S2 is output as a code word code (ω), and a new reference number is added to the character string (ωK) and registered in the dictionary. The input character K of is replaced with the reference number ω, the dictionary address N is incremented, and after the check in step S6 is received, the process returns to step S2 to read the next character K.

Next, the encoding will be specifically described with reference to FIGS. 11 and 12. Incidentally, FIG. 11 and FIG.
In order to simplify the explanation, the case of compressing data consisting of a combination of three letters abc is taken up. First, FIG.
Input data of is read from left to right. When the first character a is input, since there is no matching character string other than a in the dictionary, the output code (reference number ω) is output as a code word.
Then, the reference character 4 is attached to the expanded character string ab and registered in the dictionary. The actual registration is in the form of a character string (1b).

Subsequently, the second b becomes the beginning of the character string.
Since there is no matching character string other than b in the dictionary, the reference number 2 is output as a codeword, and the expanded character string ba is actually added to the dictionary with the reference number 5 and registered in the dictionary. The third a is the beginning of the next character string. Hereinafter, this processing is similarly continued. The decoding process of FIG. 10 performs the reverse operation of the encoding process of FIG.

In the decoding of FIG. 10, similarly to the encoding in step S1, a character string consisting of one character for every character is registered in the dictionary in advance as an initial value, and then the decoding is started. First, in step S2, the first code (reference number) is read,
The current CODE is set as an OLD code, and the first code corresponds to any one-character reference number already registered in the dictionary. Therefore, the character code (K) matching the input code CODE is searched for, and the character K is output. The output character (K) is set in char for exception processing later.

Next, in step S3, the next code is read and set in CODE as NEW code. Next, in step S4, the code CO input in step S3 is input.
It is checked whether DE is defined (registered) in the dictionary. Normally, the input codeword is registered in the dictionary by the processing up to the previous time, so the processing proceeds to step S5 and the code CODE is entered.
The character string code (ωK) corresponding to is read from the dictionary, the character string K is temporarily stacked in step S6, and the reference number
The code (ω) is set as a new CODE and returned to step S5 again, and the procedure of steps S5 and S6 is recursively repeated until the reference number ω reaches one character, and finally, the process proceeds to step S7 and the characters stacked in step S6. Is popped up and output in the LILO (Last In Fast Out) format.

At the same time, in step S7, a new reference number is added to the character string in which the code ω used last time and the first character K of the character string restored this time are represented as a set (ω, K) and registered in the dictionary. .. If the code that is not registered in step S4 (which occurs when the immediately preceding reference number is referred to in encoding), the OLDco
de to CODE, code (OLDcode, char) to NEWco
After returning to de, the process proceeds to step S5.

The decoding process will be described in detail with reference to FIG. Note that, in FIG. 1312, the case of decoding data consisting of a combination of three characters of abc is illustrated for the sake of simplicity. First, in FIG. 13, the first input code is 1, and the characters a, b, and c are already registered in the dictionary as reference numbers 1, 2, and 3 as shown in FIG. Is replaced with a character string a having a reference number that matches

Similarly, the following code 2 is replaced with the character b and output. At this time, a new reference number 4 is added to the combination (1b) of the previously processed code and the first character b decoded this time, and registered in the dictionary. The third code 4 replaces 1b with ab by searching the dictionary and outputs the character string ab. At the same time, a character string 2a obtained by combining the previously processed code 2 and the first character a of the character string decoded this time
(= Ba) is added to the reference number 5 and registered in the dictionary. Similarly, this process is repeated thereafter.

However, there is the following exception processing in the decoding of FIG. This exception processing occurs in the decoding of the sixth input code 8. Code 8 is not defined in the dictionary at the time of decoding,
I can't decrypt. In this case, the character string 5b obtained by adding the first character b of the previously decoded character string ba to the code 5 processed last time is obtained, and further replaced with 2ab and bab and output. Then, the reference number 8 is added to the character string 5b obtained by adding the character b of the character string decoded this time to the previous code word 5 to the output word of the character string and registered in the dictionary.

This exception processing is performed through the processing of steps S4 and S9 of the decoding processing flow of FIG.
Finally, in step S7, the character string is output and the new character string is added to the reference number and registered in the dictionary. The encoding and decoding processes of FIGS. 9 and 10 are performed while creating the same dictionary.

[0018]

In the conventional LZW code, when the input character code and the data are divided into different character strings for encoding, each character string currently being encoded is independent of the previous character string. It takes the form of encoding as it appears in. With this method, there is no problem in encoding the memoryless information source.

However, many data such as actual sentences are regarded as a memory information source, and the history of appearance of character strings cannot be fully utilized in LZW encoding, and the dependency of appearance of character strings even after data compression. There was a drawback that the redundancy remained. FIG. 14 is a search tree of the dictionary created when the LZW encoding of FIG. 9 is performed, the root of the search tree of the dictionary is empty, and in the LZW code, it appears before the character string being encoded. The history of the string that was done is not considered. Further, FIG. 15 shows the LZW code generated along with the creation of the dictionary of FIG.

Therefore, the inventors of the present invention use the dictionary to determine the subordinate relationship with the last character group of the immediately preceding character string (including the character before the last character, the character two before the last character ...). We have proposed an LZW encoding and decompression method that reduces redundancy between character strings by incorporating it and increases the compression rate. Specifically, as shown in FIG. 16, a plurality of dictionaries, for example, 0
It is possible to select an individual dictionary by dividing each of the characters up to 255 and indexing them, and indexing the last characters P0, P1, P2, P3, ... Of the immediately preceding character string as shown in FIG. In each dictionary, the last characters P0, P1, P2, P3, ...
・ Store only the character string connected to.

According to this method, conventionally, a code word is represented by a reference number of a character string in the dictionary as a whole, but the code word can be represented by a reference number of a series connected to an index, which is short. It can be expressed and the coding efficiency can be improved. However, this method has the following problems. When dividing into a plurality of dictionary capacities, for example, when divided into 256, the capacity of 1/256 (equal division) of the entire dictionary capacity is assigned to each divided dictionary in the initial state, and therefore the dictionary is not accessed at all. Although the same capacity is allocated to, the frequently used dictionaries have a drawback that the dictionary capacity soon exceeds.

In order to use the entire dictionary capacity more efficiently, including the registration of the initial value, the history state of the dependency relation with the last character is degenerated (for example, the history state of only the upper bit of the last character is changed). It may be possible to use a large capacity of the dictionary that is often used, but without knowing the characteristics of the data in advance, the dictionary capacity cannot be distributed, and there is little effect on arbitrary data. It had drawbacks. The present invention has been made in view of such a conventional problem, and when the encoding and decoding are performed by designating a division dictionary as the last character group of the immediately preceding character string as history information, the capacity of the division dictionary is reduced. It is an object of the present invention to provide a data compression and decompression method that can appropriately allocate and reduce the memory capacity used as a dictionary and can suppress an unnecessary increase in the reference number of the dictionary to obtain a high compression rate.

[0023]

FIG. 1 illustrates the principle of the present invention. First, the present invention is based on the dictionary 10 having a finite size.
The target is a data compression method in which an input character string is encoded by an encoded character substring registered in. According to the present invention as such a data compression method, FIG.
, The dictionary 10 is divided into a plurality of divided dictionaries 10-0 to 10-n, and the divided dictionaries 10-1 to 10-10.
The initial setting means 12 for setting the minimum dictionary size for -n, and the last character group in the character string encoded immediately before the encoding of the input character string, for example, specified by the history based on the last character The dictionary searching means 14 for designating the divided dictionary 10-i of the above and searching for a substring having the maximum length match among the already encoded substrings registered in the designated divided dictionary 10-i.
An encoding unit 16 for outputting, as a code word, a reference number of a subsequence having a maximum length matching the input character retrieved by the dictionary retrieval unit 14 from the division dictionary 10-i; and the division dictionary 10-i used for encoding. If the reference number exceeds the predetermined dictionary size, the dictionary capacity of the divided dictionary 10-i is increased, and if it does not exceed the dictionary size, the dictionary capacity increasing means 18 for maintaining the dictionary size as it is and the encoding means 16 store the code word. A division dictionary registration means 20 is provided for registering a character string obtained by adding the next input character to a code word in a division dictionary designated by a history based on the last character group of the code word when output. And

The present invention also provides a dictionary 1 having a finite size.
The data restoration method for decoding the input codeword string by the decoded character substring registered in 0 is targeted. According to the present invention regarding this data restoration method, FIG.
As shown in, the initial setting means 12 for setting the minimum dictionary size for each divided dictionary, and the last character group in the character string decoded immediately before the decoding of the input codeword, for example, the last character. According to the history based on the division dictionary 10-0 ~ 1
A character or character string based on the division dictionary search means 14 for searching a specific division dictionary 10-i from 0-n and the registered content of the reference number that matches the input codeword of the retrieved division dictionary 10-i. If the reference size of the divided dictionary 10-i used for decoding is larger than the predetermined dictionary size, the dictionary capacity of the divided dictionary 10-i is increased. When the character string is decoded by the dictionary capacity increasing unit 18 that maintains the dictionary size as it is and the decoding unit 22, the character string obtained by adding the first character of the restored character string to the previous input codeword is immediately preceding. And a division dictionary registration means 24 for registering in the division dictionary designated by the history based on the final character group of the decoded character string.

Here, the initial setting means 12 uses the divided dictionaries 1
Every 0-0 to 10-n at least all character types, for example, 25
All six character types are assigned a reference number on a character-by-character basis to set a minimum dictionary size that can be registered, and all the character types are assigned a reference number on a character-by-character basis for initial registration. Here, the dictionary capacity increasing means 18 sets the reference number of the divided dictionary used for encoding or decoding each time it exceeds a predetermined dictionary size.
Increase the capacity of the split dictionary to twice the current size.

Further, the dictionary capacity increasing means 18 increases the dictionary capacity to a size obtained by adding a predetermined minimum unit to the current size each time the reference number of the divided dictionary used for encoding or decoding exceeds a predetermined dictionary size. It may be allowed to.

[0027]

According to the data compression and decompression method of the present invention having such a configuration, the following effects can be obtained. First, in the data compression method of the present invention, as shown in the flowchart of FIG. 2, first, a minimum dictionary size is given to each divided dictionary, and then the same character as the input data is set according to the history between character strings. Searches from the divided dictionaries, then encodes with the index (reference number) in the divided dictionary, and if the index (reference number) of the divided dictionary used for encoding exceeds the predetermined dictionary size, divides By increasing the dictionary capacity of the dictionary and registering the dictionary size as it is when the dictionary size is not exceeded, a finite dictionary can be used efficiently and the frequently used division dictionary is prevented from exceeding.

Further, in the data restoration system of the present invention, as shown in the flowchart of FIG. 3, first, a minimum dictionary size is given to each divided dictionary, and then the input coded data (code word) is used. The corresponding character string is searched from the dictionary divided according to the history between the character strings, and the character string detected from the divided dictionary is restored. At this time, if the index (reference number) of the divided dictionary used for decoding exceeds the predetermined dictionary size, increase the dictionary capacity of the divided dictionary, and if it does not exceed the dictionary size, register the dictionary size as it is. Similarly, a finite dictionary is used efficiently so that the frequently used division dictionary is not overrun.

The increase in the size of the division dictionary in the data encoding and decoding is performed by giving the minimum size to each division dictionary in the first step, as shown in FIG. 4, for example.
The initial value is registered, and the final absolute address amax (n) of each divided dictionary is set to n × 256 and the maximum dictionary capacity dmax (n) is set to 256 and stored in another area. Also, the final absolute address all-amax of the whole dictionary is 66536, the maximum dictionary capacity of the whole dictionary.
Set all-dmax to 66536.

In the second stage, if a divided dictionary with n = 0 is selected, the final absolute address all-
The area from amax to 256 is secured in the divided dictionary of n = 0, and the final absolute address amax (0) of each divided dictionary and the maximum dictionary capacity dma
The final absolute address all-amax and the maximum dictionary capacity all-dmax of x (0) and the whole dictionary are updated as follows. The third stage is a case where the division dictionary of n = 256 exceeds the current size (minimum size) 256, and is doubled to 512.

In the fourth step, when the division dictionary of n = 0 exceeds the current size 512, it is doubled to 1024. In the fifth step, when the division dictionary of n = 0 exceeds the current size 1024, it is doubled to 2048. This is summarized as follows. [Last absolute address] [Maximum dictionary capacity] [Last absolute address] [Maximum dictionary capacity] 1st stage: n × 256 → amax (n), 256 → dmax (n), 66536 → all-amax, 66536 → all- dmax 2nd stage: 66792 → amax (0), 512 → dmax (0), 66792 → all-amax, 66792 → all-dmax 3rd stage: 67048 → amax (255), 512 → dmax (255), 67048 → all-amax, 67048 → all-dmax 4th stage: 67560 → amax (0), 1024 → dmax (0), 67560 → all-amax, 67560 → all-dmax 5th stage: 68584 → amax (0), 2048 → dmax (0), 68584 → all-amax, 68584 → all-dmax As described above, the limited dictionary capacity is used as effectively as possible according to the data. However,
In the case of FIG. 4, the dictionary capacity is updated to twice the current size, but the dictionary capacity to be added is arbitrary.

[0032]

FIG. 5 is a block diagram of an embodiment showing one embodiment of the present invention. In FIG. 5, reference numeral 26 is a CP as a control means.
U, the program memory 28 for the CPU 26
And the data memory 32 are connected. Program memory 2
8, control software 30, initial setting software 1
2, dictionary search software 14, encoding software 16, decoding software 22, dictionary capacity increasing software 18, and divided dictionary registration software 20, 24 are provided.

The initialization software 12 is a division dictionary 10-1 to 10 divided into, for example, 256 when encoding and decoding.
-Set the minimum size for each of -255,
All the character types 256 are initially registered by indexing each character. When the input character string is encoded, the dictionary search software 14 designates a specific divided dictionary 10-i based on the last character group in the character string encoded immediately before, for example, the history based on the character code of the last character, Of the already encoded substrings registered in the designated division dictionary 10-i, the substring having the maximum length match is searched.

Further, the dictionary search software 14 divides the input coded data (codeword) by the history based on the last character group in the character string decoded just before, for example, the character code of the last character, at the time of decoding. A specific divided dictionary 10-i is searched from the dictionaries 10-0 to 10-n. Encoding software 16
Outputs, as encoded data (code word), a reference number indicating a substring having the maximum length that matches the input character of the already encoded substring searched by the dictionary search software 14 from the divided dictionary.

The decoding software 22 also decodes a character or a character string based on the registered content of the reference number that matches the input codeword of the divided dictionary 10-i searched by the dictionary search software 14. The dictionary capacity increasing software 18 increases the dictionary capacity of the divided dictionary 10-i when the index of the divided dictionary 10-i used for encoding or decoding exceeds a predetermined dictionary size, and when it does not exceed the dictionary size, the dictionary capacity is increased. Keep the size as it is. Further, when the encoding software 16 outputs the code word, the division dictionary registration software 20 specifies the character string obtained by adding the next input character to the code word by the history based on the last character group of the code word. Register in the split dictionary. Further, when the decoding software 22 decodes the character string, the division dictionary registration software 24 adds a character string obtained by adding the leading one character of the restored character string of this time to the character string of the character string decoded immediately before. It registers in the division dictionary designated by the history based on the last character group.

On the other hand, in the data memory 32, a data buffer 34 for storing a character string to be encoded or a code string to be decoded, and 256 divisions corresponding to all 256 character types to be processed. The divided dictionaries 10-0 to 10-255 are provided. FIG. 6 is a flowchart showing details of the encoding process of the present invention.

First, in step S1, as a default, the total number M of appearing characters, for example M = 256, the number A of divided dictionaries, for example A = 256, is set to A = 256.
Initially register M = 0 to 255 characters in this divided dictionary Di. Next, the number of nodes (indexes) is managed by indc (i) for each tree i of the A division dictionaries selected by the last character of the immediately preceding character string. First, as initialization, A indcs
Set (i) to M + 1 (= 256). Further, the maximum dictionary capacity dmax (i) of each divided dictionary is set to the minimum dictionary size M + 1.
(= 256).

First, the first character K is input, and it is used as an index (initial character string) ω and also substituted for the last character K1 of the immediately preceding character string. A lookup table LU that defines the history PK from the last character of the immediately preceding character string and associates it with the dictionary number to be used from the last character K1 of the immediately preceding character string.
Install T. Since there is no last character K1 of the immediately preceding character string for the first character, a predetermined fixed value is set in the history PK.
To use as.

Next, in step S2, the next character K is input. Then, in step S3, it is checked whether or not the character string ωK exists in the division dictionary D _PK . If it exists, the process proceeds to step S4, the character string ωK is replaced with a new initial character string ω, the input character K is replaced with the final character K1, and the process returns to step S2 via step S5 to search for the longest matching character string. To do.

If the character string ωK does not exist in the divided dictionary D _PK in step S3 and the search for the longest character string is completed, the process proceeds to step S6. In step S6, the index indc split dictionary D _PK (PK) is, dictionary size dmax of split dictionary D _PK (PK) to check whether larger. The index indc (PK) is the dictionary size dmax (P
If it exceeds K), the process proceeds to step S7 and the dictionary size dm
The ax (PK) and the final absolute address amax (PK) are updated, and the process proceeds to step S8.

In step S6, the index indc
If (PK) does not exceed the dictionary size dmax (PK), the process proceeds to step S8. In step S8, the encoded data code (ω) of the divided dictionary is output and the address indc
After the character string ωK is registered in the divided dictionary D _PK of (PK), the character K is substituted for the initial character string ω, and the index indc (P
K) is incremented, the history PK is set to LUT (K1), and the process proceeds to step S5.

FIG. 7 is an explanatory diagram concretely showing the increase of the capacity of the division dictionary performed in the encoding process of FIG. 6, and corresponds to FIG. In FIG. 7, a first step of initial setting and initial registration; a second step of securing a dictionary area when a divided dictionary of n = 0 is selected; a dictionary when a divided dictionary of n = 256 is selected Third stage of securing a region; Fourth stage of securing a further dictionary region when the size of the divided dictionary of n = 0 is exceeded; Fourth stage of securing a further dictionary region when the size of the divided dictionary of n = 0 is exceeded Stage;

That is, in the first stage, the minimum size is given to each divided dictionary, the initial value is registered, the final absolute address amax (n) of each divided dictionary is n × 256, and the maximum dictionary capacity dmax.
Set (n) to 256 and store it in another area. Also, the final absolute address all-amax of the whole dictionary is set to 66536 and the maximum dictionary capacity all-dmax of the whole dictionary is set to 66536. At the second stage, since the divided dictionary with n = 0 is selected, the area from the final absolute address all-amax of the entire dictionary to 256 is n.
The final absolute address amax (0) and the maximum dictionary capacity dmax (0) of each divided dictionary and the final absolute address all-amax and the maximum dictionary capacity all-dmax of the whole dictionary are updated.

In the third step, when the number of divided dictionaries of n = 256 exceeds the current size (minimum size) 256, 2
Doubled to 512 and the final absolute address of each split dictionary
Update amax (0) and maximum dictionary capacity dmax (0), and final absolute address all-amax and maximum dictionary capacity all-dmax of the whole dictionary. In the fourth stage, when the division dictionary of n = 0 exceeds the current size 512, it is doubled to 1024.

In the fourth stage, when the division dictionary of n = 0 exceeds the current size 1024, it is doubled to 2048. FIG. 8 is a flowchart showing details of the decoding process of the present invention. First, the initial setting of step S1 is the same as step S1 of FIG. Then, in step S2, the first code is read and set as an OLD code.
Restore the character K from the split dictionary D _PK corresponding to CODE,
Outputs the character K to char, PK to PK1, LU
Substitute T (K) into PK. Next, in step S3, the next code is read and set as NEW code.

In step S4, the COD is added to the division dictionary D _PK .
If E is not defined, the process proceeds to step S5, and if it is defined, the process proceeds to step S6. In step S5, the first character char of the previous character string is output and the COD
E is returned to OLD code, and NEW code is divided dictionary D _PK
After returning to the code obtained from the combination of OLD code and char in the above, the process proceeds to step S6.

In step S6, the division dictionary D _PK
Character string code (ωK) corresponding to the index CODE of
From the dictionary, the character string K is temporarily stacked in step S7, the reference number code (ω) is returned to step S6 as a new CODE, and the procedure of steps S6 and S7 is recursively performed. Is repeated until one character is reached, and finally the process proceeds to step S8 and step S8.
Characters stacked in 7 are LILO (Last In Fast Out)
Pop up in a format and output.

Next, in step S9, the division dictionary D _PK
, Index indc (PK) is larger than the dictionary size dmax (PK) of the division dictionary D _PK .
If the index indc (PK) exceeds the dictionary size dmax (PK), the process proceeds to step S10, and the dictionary size dmax
(PK) and the final absolute address amax (PK) are updated.

If the index indc (PK) does not exceed the dictionary size dmax (PK), the process proceeds to step S11. In step S11, a combination address indc (PK1) of the immediately preceding code OLDcode and the last character K of the immediately preceding character string
Register in the divided dictionary D _PK . Then the index indc (P
The value of K1) is incremented, and the process proceeds to step S12. In step S13, the first character of the restored character string is cha
r, the last character of the restored character string is K1, the history PK is PK1
, LUT (K1) to PK, NEWcode to OLDcode
, And returns to step S3 via step S14.

Also in this decoding process, the capacity of the divided dictionary similar to that shown in FIG. 7, for example, is added. It should be noted that the addition of the capacity of the divided dictionary in the above-described embodiment is performed by doubling the current size, which is 256, 512, 1024,
Although the case of increasing the number to 48, ... Is taken as an example, the dictionary size to be increased is set to, for example, 256 at the time of initial setting.
Fixed to 256, 512, 768, 1024, ...
The method of increasing the dictionary size can be appropriately determined as necessary.

[0051]

As described above, according to the present invention, the capacity of the divided dictionary is increased according to the number of registrations at the time of encoding or decoding, so that the dictionary capacity as a whole is effectively used. Since a code word with a short reference number (index) can be used for any data, a high compression rate can be expected.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of the principle of the present invention.

FIG. 2 is an operation explanatory view showing an encoding process of the present invention.

FIG. 3 is an operation explanatory view showing a decoding process of the present invention.

FIG. 4 is an operation explanatory view showing an increase in dictionary capacity according to the present invention.

FIG. 5 is a configuration diagram of an embodiment of the present invention.

FIG. 6 is a flow chart showing details of encoding of the present invention.

7 is an explanatory diagram showing a situation in which the dictionary capacity of a divided dictionary is added in the encoding of FIG.

FIG. 8 is a flowchart showing details of the decoding of the present invention.

FIG. 9 is a flowchart showing conventional LZW encoding.

FIG. 10 is a flowchart showing conventional LZW decoding.

FIG. 11 is an explanatory diagram showing a specific example of conventional LZW encoding.

FIG. 12 is an explanatory diagram specifically showing a dictionary structure created by conventional LZW encoding.

FIG. 13 is an explanatory diagram showing a specific example of conventional LZW decoding.

FIG. 14 is a tree structure diagram of a dictionary in a conventional LZW code.

FIG. 15 is an explanatory diagram showing a relationship between character strings by conventional LZW encoding.

FIG. 16 is a tree configuration diagram of a dictionary of a split dictionary system proposed by the inventors of the present application.

FIG. 17 is an explanatory diagram showing a relationship between character strings encoded by the division dictionary method proposed by the inventors of the present application.

[Explanation of symbols]

 10: Dictionary 10-0 to 10-255: Divided dictionary 12: Initial setting means (initial setting software) 14: Dictionary search means (dictionary search software) 16: Encoding means (encoding software) 18: Dictionary capacity increasing means ( Dictionary capacity increasing software) 20, 24: Divided dictionary registration means (divided dictionary registration software) 22: Decoding means (decoding software) 26: CPU 28: Program memory 30: Control software 32: Data memory 34: Data buffer

 ─────────────────────────────────────────────────── --- Continuation of the front page (72) Inventor Hirotaka Chiba 1015 Kamiodanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture Fujitsu Limited

Claims (5)

[Claims] 1. A data compression method for encoding an input character string by an encoded character substring registered in a dictionary (10) having a finite size, wherein a plurality of the dictionary (10) are provided. A divided dictionary (10-0 to 10-n) configured by dividing, initial setting means (12) for setting a minimum dictionary size for each divided dictionary (10-1 to 10-n), and input characters At the time of encoding a string, a specific division dictionary (10-i) is designated by a history based on the last character group in the character string that has been encoded immediately before, and is registered in the designated division dictionary (10-i). The dictionary search means (14) for searching a substring having the maximum length match among the already encoded substrings, and the maximum length match with the input character string searched from the divided dictionary by the dictionary search means (14) An encoding means (16) for outputting the reference number of the subsequence as a codeword; If the reference number of the division dictionary (10-i) used for the encoding exceeds a predetermined dictionary size, the division dictionary (10-i)
The dictionary capacity increasing means (18) for increasing the dictionary capacity of -i) and maintaining the dictionary size as it is when it does not exceed, and when the code word is output by the encoding means (16), A data compression method, comprising: a division dictionary registration means (20) for registering a character string to which one input character is added in a division dictionary specified by a history based on the last character group of the codeword. 2. A data restoration method for decoding a codeword string input by a decoded character substring registered in a dictionary (10) having a finite size, wherein a plurality of said dictionary (10) are provided. Divided dictionary (10-0 to 10-n) configured by dividing the divided dictionary, initial setting means (12) for setting a minimum dictionary size for each divided dictionary, and immediately before decoding the input codeword. A dictionary based on the history based on the final character group in the decoded character string (10-
0 to 10-n) to search for a specific divided dictionary (10-i), and a reference number that matches the input codeword of the searched divided dictionary (10-i). Decoding means (22) for decoding a character or a character string based on the registered contents, and the division dictionary (10-i) used for the decoding when the reference number exceeds a predetermined dictionary size, the division dictionary. (10
The dictionary capacity increasing means (18) for increasing the dictionary capacity of -i) and maintaining the dictionary size as it is when it does not exceed, and the previous input codeword when the character string is decoded by the decoding means (22). A division dictionary registration means (24) for registering a character string obtained by adding the first character of the restored character string to a division dictionary designated by a history based on the last character group of the character string decoded immediately before is provided. Data restoration method characterized by 3. The data compression and decompression method according to claim 1 or 2, wherein said initial setting means (12) is provided for each division dictionary (10-0 to 0-0).
10-n) at least all character types are assigned a reference number in a character unit to set a minimum dictionary size that can be registered, and all character types are initially registered in a character number with a reference number. Data compression and decompression method. 4. The data compression and decompression method according to claim 1 or 2, wherein the dictionary capacity increasing means (18) has a predetermined reference number of a division dictionary used for encoding or decoding. A data compression and decompression method characterized by increasing the capacity of the division dictionary to twice the current size each time the size is exceeded. 5. The data compression and decompression system according to claim 1 or 2, wherein the dictionary capacity increasing means (18) has a predetermined reference number of a divided dictionary used for encoding or decoding. A data compression and decompression method characterized by increasing the dictionary capacity to a size obtained by adding a predetermined minimum unit to the current size each time the size is exceeded.