Data compressing apparatus, data decompressing apparatus, data compressing method, data decompressing method, and program recording medium

• US 6,094,634 A
• Filed: 01/23/1998
• Issued: 07/25/2000
• Est. Priority Date: 03/26/1997
• Status: Expired due to Term

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Yahagi, Hironori, Morihara, Takashi
• Primary Examiner(s): Hudspeth, David R.
• Assistant Examiner(s): Lerner, Martin
• Application Number:
  US09/012,548
• Time in Patent Office: 914 Days
• Field of Search: 704/201, 704/235, 704/260, 704/500, 704/503, 704/8, 704/2, 704/7, 341/50, 341/51, 341/90
• US Class Current: 704/260
• CPC Class Codes: G06F 40/53 Processing of non-Latin text kana-to-kanji conversion G06F40/129; vowelisation G06F40/232
• CPC Class Codes: H03M 7/30 Compression speech analysis-synthesis for redundancy reduction G10L19/00; for image communication H04N; Expansion; Suppression of unnecessary data, e.g. redundancy reduction

Abstract

A data compressing/decompressing apparatus is suitable for compressing data containing plural-byte characters, for instance, a Japanese language text. The data compressing/decompressing apparatus owns a homonym dictionary in which KANJI-character idioms, KANJI-character-reading, and homonym discrimination information are stored in correspondence with each other. This data compressing/decompressing apparatus converts a KANJI-character idiom contained in declarative sentence data into phonetic data, and further compresses this phonetic data to output the compressed phonetic data. The phonetic data is such data that this KANJI-character idiom is replaced by information constituted by character number discrimination information indicative of the character number about KANJI-character idiom reading, the reading thereof, and the homonym discrimination information thereof.

Claims

1. A data compressing apparatus for compressing original text data containing character information in which a single character is expressed by plural-byte information, comprising:

phonetic text data producing means for producing phonetic text data equal to the original text data such that the character information contained in the plural-byte coded original text data is replaced by phonetic character information, expressed as a phonogram in 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in the original text data; and

compressing means for compressing the phonetic text data produced by said phonetic text data producing means.

2. A data compressing apparatus for compressing original text data containing character information in which a single character is expressed by plural-byte information, comprising:

phonetic character information storing means for storing phonetic character information expressed with phonograms in 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in the original text data and equivalent to said character information;

retrieving/reading means for retrieving the word information to be converted stored in said phonetic character information storing means from said original text data, and also for reading phonetic character information corresponding to the retrieved word information to be converted from said phonetic character information storing means;

phonetic text data producing means for producing phonetic text data by replacing the word information to be converted within said plural-byte coded original text data retrieved by said retrieving/reading means with word replacement information to be converted, which contains the phonetic character information read by said retrieving/reading means;

intermediate code table forming means for forming an intermediate code table used to adapt an intermediate code to an information element utilized in the phonetic text data produced by said phonetic text data producing means;

intermediate code text data by producing means for producing intermediate code text data by converting the respective information elements for constructing said phonetic text data into the corresponding intermediate codes by using the intermediate code table produced by said intermediate code producing means; and

compressing means for compressing the intermediate code text data produced by said intermediate code text data producing means.

3. A data compressing apparatus as claimed in claim 2 wherein:

said intermediate code table forming means forms an intermediate code table for allocating an intermediate code having a minimum bit number to each of the information elements used in said phonetic text data, and said minimum bit number is capable of expressing these information elements.

4. A data compressing apparatus as claimed in claim 2 wherein:

said phonetic text data producing means produces the phonetic text data by replacing a preselected sort of character information by replacement information sandwiched between starting position discrimination information and end position discrimination information, which indicate that said character information is equal to the preselected sort of character information;

said intermediate code table forming means forms such an intermediate code table as the intermediate codes do not correspond to both information subsequent to said starting position discrimination information and said end position discrimination information within said phonetic text data; and

said intermediate code data producing means does not convert both the information subsequent to said starting position discrimination information and said end position discrimination information within said phonetic text data into the intermediate codes.

5. A data compressing apparatus as claimed in claim 2 wherein:

said intermediate code table forming means selects "N-1" pieces of information elements when a sort of information elements used in said phonetic text data exceeds a number "N" of information capable of being expressed by a predetermined bit number to thereby allocate intermediate codes among the information elements used in said phonetic text data; and

also form an intermediate code table by which said predetermined bit number of intermediate codes, whose contents are different from each other, are related both to said selected "N-1" pieces of information elements and to said starting position discrimination information; and

said intermediate code data producing means produces the intermediate code text data in such a manner as with regard to the information contained within said phonetic text data, to which the intermediate code is related by said intermediate code table; and

with regard to unallocated information where no intermediate code is allocated, said unallocated information is replaced by unallocated replacement information, said unallocated replacement information being equal to such information containing the unallocated information and having an intermediate code related to said starting position discrimination information, and also capable of recognizing an end position.

6. A data compressing apparatus as claimed in claim 2 wherein:

said phonetic character information storing means stores phonetic character information and homonym discrimination information for discriminating other words to be converted to which the same homonym discrimination character information is adapted with respect to said plurality of words to be converted;

said retrieving/reading means reads out the phonetic character information and the homonym discrimination information, which correspond to the retrieved words to be converted;

said phonetic text data producing means replaces the word information to be converted within said original text data by word replacement information to be converted which contains the phonetic character information and the homonym discrimination information read by said retrieving/reading means; and

said intermediate code table forming means forms the intermediate code table directed to the information elements except for said homonym discrimination information.

7. A data compressing apparatus as claimed in claim 2 wherein:

said phonetic character information storing means stores phonetic character information and homonym discrimination information for discriminating other words to be converted to which the same homonym discrimination character information is adapted with respect to said plurality of words to be converted;

said retrieving/reading means reads out the phonetic character information and the homonym discrimination information, which correspond to the retrieved words to be converted; and

said phonetic text data producing means replaces the word information to be converted within said original text data by word replacement information to be converted which contains the phonetic character information and the homonym discrimination information read by said retrieving/reading means;

said word replacement information containing the homonym discrimination information at a specific position.

8. A data compressing apparatus as claimed in claim 2 wherein:

said phonetic text data producing means applies numeral value discrimination information indicative of a length of the phonetic character information read by said retrieving/reading means to a head portion of said word replacement information to be converted.

9. A data compressing apparatus as claimed in claim 2 wherein:

said phonetic text data producing means produces the phonetic text data by replacing said word information to be converted by such word replacement information to be replaced, which is sandwiched between phonetic character information starting position discrimination information and phonetic character information end position discrimination information, which indicate a start and an end of the word information to be converted.

10. A data compressing apparatus for compressing original text data containing character information in which a single character is expressed by plural-byte information, comprising:

phonetic character information storing means for storing phonetic character information expressed with phonograms in 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in the original text data and equivalent to said character information;

retrieving/reading means for retrieving the word information to be converted stored into said phonetic character information storing means from said original text data, and also for reading phonetic character information corresponding to the retrieved word information to be converted from said phonetic character information storing means;

phonetic text data producing means for producing phonetic text data by replacing the word information to be converted within said plural-byte coded original text data retrieved by said retrieving/reading means by word replacement information to be converted, which contains the phonetic character information read by said retrieving/reading means; and

compressing means for compressing the phonetic text data produced by said phonetic text data producing means.

11. A data decompressing apparatus comprising:

decompressing means for decompressing compression text data; and

original text data producing means for producing original text data equal to data for constituting an original of said compression text data by converting phonetic character information, restored by said decompressing means, expressed with phonograms composed of 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in decompressed text data, into character information corresponding thereto.

12. A data decompressing apparatus comprising:

phonetic character information storing means for storing phonetic character information equal to information expressed with phonograms in 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in decompressed text data and equivalent to word information to be converted being constructed of one or plural character information;

decompressing means for decompressing compression text data to output intermediate code text data;

phonetic text data producing means for producing phonetic text data by replacing each of intermediate codes contained in the intermediate code text data outputted by said decompressing means by information adapted to the intermediate code in an intermediate code table related to said compression text data; and

original text data producing means for producing original text data equal to an original of said compression text data such that word replacement information to be converted is retrieved which is contained in the phonetic text data produced by said phonetic text data producing means, and said retrieved word replacement information to be converted is replaced by word information to be converted which is stored in said phonetic character information storing means in correspondence with said phonetic character information contained in said word replacement information to be converted.

13. A data decompressing apparatus as claimed in claim 12 wherein:

as to information sandwiched between end position discrimination information and an intermediate code corresponding to starting position discrimination information outputted by said decompressing means, said phonetic text data producing means directly uses said information as an element of the phonetic text data without carrying out the replacement by employing said intermediate code table.

14. A data decompressing apparatus as claimed in claim 13 wherein:

said intermediate code table is a code table such that a preselected bit number of intermediate codes having contents different from each other are adapted to "N-1" pieces of information elements and the starting position discrimination information; and

as to unallocated replacement information starting with the intermediate code adapted to said starting position discrimination information contained in the intermediate code text data outputted by said decompressing means, said phonetic text data producing means outputs such information that another information added in order to discriminate said intermediate code from the end position thereof is removed from this unallocated replacement information.

15. A data decompressing apparatus as claimed in claim 12 wherein:

said phonetic character information storing means stores phonetic character information and homonym discrimination information for discriminating other words to be converted to which the same homonym discrimination character information is adapted with respect to said plurality of words to be converted; and

original text data producing means operated such that word replacement information to be converted is retrieved which is contained in the phonetic text data produced by said phonetic text data producing means, and said retrieved word replacement information to be converted is replaced by word information to be converted which is stored in said phonetic character information storing means in correspondence with said phonetic character information contained in said word replacement information to be converted.

16. A data decompressing apparatus as claimed in claim 12 wherein:

said phonetic character information storing means stores therein the phonetic character information and the homonym discrimination information used to discriminate the word to be converted from another word to be converted, having the same phonetic character information with respect to each of said plural word information to be converted; and

original text data producing means produces original text data equal to an original of said compression text data by such that word replacement information to be converted is retrieved which is contained in the phonetic text data produced by said decompressing means, and said retrieved word replacement information to be converted is replaced by word information to be converted which is stored in said phonetic character information storing means in correspondence with said phonetic character information contained in said word replacement information to be converted.

17. A data decompressing apparatus as claimed in claim 12 wherein:

when said word replacement information to be converted is retrieved, said original text data producing means retrieves such information starting with numeral value discrimination information indicative of a length of the phonetic character information.

18. A data decompressing apparatus as claimed in claim 12 wherein:

when said word replacement information to be converted is retrieved, said original text data producing means retrieves information sandwiched between the phonetic information starting position discrimination information and the phonetic information end position information.

19. A data decompressing apparatus comprising:

phonetic character information storing means for storing phonetic character information equal to information expressed with phonograms in 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in decompressed text data and equivalent to word information to be converted being constructed of one or plural character information;

decompressing means for decompressing means for decompressing compression text data to output phonetic text data; and

original text data producing means for producing original text data equal to an original of said compression text data such that word replacement information to be converted is retrieved which is contained in the phonetic text data produced by said decompression means, and said retrieved word replacement information to be converted is replaced by word information to be converted which is stored in said phonetic character information storing means in correspondence with said phonetic character information contained in said word replacement information to be converted.

20. A data compressing method for compressing original text data containing character information in which a single character is expressed by plural-byte information, comprising:

a phonetic text data producing step for producing phonetic text data equal to such data that each of the character information contained in the plural-byte coded original text data to be compressed is replaced by phonetic character information expressed with phonograms in 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in the original text data; and

a compressing step for compressing the phonetic text data produced by said phonetic text data producing step.

21. A data compressing method form compressing original text data containing character information in which a single character is expressed by plural-byte information, comprising:

a retrieving/reading step for retrieving word information to be converted from the plural-byte coded original text data to be compressed by using a dictionary into which phonetic character information is stored, and said phonetic character information being equal to information expressed with phonograms in 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in the original text data and equivalent to word information to be converted which is constituted by one or a plurality of character information, and also for reading phonetic character information corresponding to the retrieved word information to be converted from said dictionary;

a phonetic text data producing step for producing phonetic text data by replacing the word information to be converted within said plural-byte coded original text data retrieved by said retrieving/reading step by word replacement information to be converted, which contains the phonetic character information read by said retrieving/reading step;

an intermediate code table forming step for forming an intermediate code table used to adapt an intermediate code to each of information elements utilized in the phonetic text data produced by said phonetic text data producing step;

an intermediate code text data producing step for producing intermediate code text data by converting the respective information elements for constructing said phonetic text data into the corresponding intermediate coded by using the intermediate code table produced by said intermediate code producing step; and

a compressing step for compressing the intermediate code text data produced by said intermediate code text data producing step.

22. A data compressing method as claimed in claim 21 wherein:

said intermediate code table forming step is a step for forming an intermediate code table for allocating an intermediate code having a minimum bit number to each of the information elements used in said phonetic text data, and said minimum bit number is capable of expressing these information elements.

23. A data compressing method as claimed in claim 22 wherein:

said intermediate code table forming step is such a step for selecting "N-1" pieces of information elements when a sort of information elements used in said phonetic text data exceeds a number "N" of information capable of being expressed by a predetermined bit number to thereby allocate intermediate codes among the information elements used in said phonetic text data; and

also for forming an intermediate code table by which said predetermined bit number of intermediate codes, whose contents are different from each other, are related to said selected "N-1" pieces of information elements and also to said starting position discrimination information; and

said intermediate code data producing means step is such a step for producing the intermediate code text data in such a manner as concerning information contained within said phonetic text data, to which the intermediate code is related by said intermediate code table; and

as to unallocated information equal to such information to which no intermediate code is allocated, said unallocated information is replaced by unallocated replacement information, said unallocated replacement information being equal to such information containing the unallocated information and having an intermediate code related to said starting position discrimination information, and also capable of recognizing an end position.

24. A data compressing method as claimed in claim 21 wherein:

said retrieving/reading step is a step for reading the phonetic character information and the homonym discrimination information, which correspond to the word to be converted, from the dictionary into which the phonetic character information and the homonym discrimination information are stored with respect to each of said plural words to be converted, said homonym discrimination information being used to be discriminated from another word to be converted, having the same phonetic character information; and

said phonetic text data producing step is a step for replacing the word information to be converted within said original text data by word replacement information to be converted which contains the phonetic character information and the homonym discrimination information read by said retrieving/reading step.

25. A data compressing method as claimed in claim 21 wherein:

said phonetic text data producing step applies numeral value discrimination information indicative of a length of the phonetic character information read by said retrieving/reading step to a head portion of said word replacement information to be converted.

26. A data compressing method for compressing original text data containing character information in which a single character is expressed in plural-byte information, comprising:

a retrieving/reading step for retrieving word information to be converted from the plural-byte coded original text data to be compressed by using a dictionary into which phonetic character information is stored, and said phonetic character information being equal to information expressed with phonograms in 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in the original text data and equivalent to word information to be converted which is constituted by one or a plurality of character information;

said word information being stored in said dictionary; and

also for reading phonetic character information corresponding to the retrieved word information to be converted from said dictionary;

a phonetic text data producing step for producing phonetic text data by replacing the word information to be converted within said plural-byte coded original text data retrieved by said retrieving/reading step by word replacement information to be converted, which contains the phonetic character information read by said retrieving/reading step; and

a compressing step for compressing the phonetic text data produced by said phonetic text data producing step.

27. A data decompressing method comprising:

a decompressing step for decompressing compression text data; and

an original text data producing step for producing original text data equal to data for constituting an original of said compression text data by converting phonetic character information, restored by said decompressing step, expressed with phonograms in 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in decompressed text data, into character information corresponding thereto.

28. A data decompressing method comprising:

a decompressing step for decompressing compression text data to output intermediate code text data;

a phonetic text data producing step for producing phonetic text data by replacing each of intermediate codes contained in the intermediate code text data outputted by said decompressing step by information adopted to an intermediate code table related to said compression text data; and

an original text data producing step for producing original text data equal to an original of said compression text data such that word replacement information to be converted is retrieved which is contained in the phonetic text data produced by said phonetic text data producing step, while using a dictionary for storing phonetic character information equal to information expressed with phonograms in 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in decompressed text data and equivalent to word information to be converted which is constituted by one or a plurality of information, and said retrieved word replacement information to be converted is replaced by word information to be converted which corresponds to said phonetic character information to be converted.

29. A data decompressing method as claimed in claim 28 wherein:

said intermediate code table is a code table such that a preselected bit number of intermediate codes having contents different from each other are adapted to "N-1" pieces of information elements and the starting position discrimination information; and

as to unallocated replacement information starting with the intermediate code adapted to said starting position discrimination information contained in the intermediate code text data outputted by said decompressing step, said phonetic text data producing step outputs such information as another information added in order to discriminate said intermediate code from the end position thereof is removed from this unallocated replacement information.

30. A data decompressing method as claimed in claim 28 wherein:

said original text data producing step replaces word replacement information to be converted which contains phonetic character information and homonym discrimination information within said phonetic text data produced by said decompressing step by the corresponding word information to be converted, while using a dictionary for storing therein the phonetic character information and the homonym discrimination information used to discriminate the word to be converted from another word to be converted, having the same phonetic character information with respect to each of said plural word information to be converted.

31. A data decompressing method as claimed in claim 28 wherein:

said original text data producing step retrieves said word replacement information to be converted by retrieving information sandwiched between starting position discrimination information and end position information.

32. A data decompressing method as claimed in claim 28 wherein:

said original text data producing step retrieves said word replacement information to be converted by retrieving information sandwiched between phonetic information starting position discrimination information and phonetic information end position information.

33. A data decompressing method comprising:

a decompressing step for decompressing compression text data to output phonetic text data; and

an original text data producing step for producing original text data equal to an original of said compression text data such that word replacement information to be converted is retrieved which is contained in the phonetic text data produced by said decompressing step, while using a dictionary for storing phonetic character information equal to information expressed with phonograms in 1-byte code characters in which the number of kinds of characters is less than the number of kinds of characters in decompressed text data and equivalent to word information to be converted which is constituted by one or a plurality of character information, and said retrieved word replacement information to be converted is replaced by word information to be converted which corresponds to said phonetic character information to be converted.

34. A program recording medium for recording there into a program capable of causing a computer to function as:

phonetic text data producing means for producing phonetic text data equal to such data as each of the character information contained in the plural-byte coded original text data to be compressed is replaced by phonetic character information expressed with phonograms in 1-byte code characters in which the number of kinds of characters in less than the number of kinds of characters in the original text data; and

compressing means for compressing the phonetic text data produced by said phonetic text data producing means.

35. A program recording medium for recording there into a program capable of causing a computer to function as:

decompressing means for decompressing compression text data; and

original text data producing means for producing original text data equal to data for constituting an original of said compression text data by converting phonetic character information as expressed with phonograms in 1-byte code characters in which the number of kinds of character is less than the number of kinds of characters in decompressed text data, as restored by said decompressing means into character information corresponding thereto in which a single character is expressed by plural-byte code character.

36. A data compressing apparatus for compressing original document data containing character information in which a single character is expressed by plural-byte code character comprising:

data converting means for converting the original document data expressed in plural-bytes code characters into document data expressed in 1-bytes code characters which designate the pronunciation of each word in the original document data; and

compressing means for compressing the document data expressed in 1-bytes code characters.