Image processing apparatus, image processing method and computer-readable medium

• US 8,411,955 B2
• Filed: 02/19/2008
• Issued: 04/02/2013
• Est. Priority Date: 02/21/2007
• Status: Expired due to Fees

• Current Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Original Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Inventors: Kimura, Shunichi, So, Ikken, Sekino, Masanori, Koshi, Yutaka
• Primary Examiner(s): Le, Vu
• Assistant Examiner(s): WOLDEMARIAM, AKLILU K
• Application Number:
  US12/071,257
• Publication Number:
  US 20080199081A1
• Time in Patent Office: 1,869 Days
• Field of Search: 382/298, 382/248, 382/233, 382/232, 382/240, 382/173
• US Class Current: 382/182
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/1823 using vector-coding
• CPC Class Codes: G06V 30/245 Font recognition

Abstract

An image processing apparatus includes an extracting unit, a representative-image generating unit and a vector converting unit. The extracting unit extracts pixel blocks from image data. The representative-image generating unit generates representative images from the pixel blocks extracted by the extracting unit, based on a similarity between the pixel blocks. The vector converting unit converts the representative images generated by the representative-image generating unit into vector information.

Claims

1. An image processing apparatus comprising:

an extracting unit that extracts pixel blocks from image data;

a representative-image generating unit i) that aligns each center of gravity of each extracted pixel block, (ii) that calculates a binary value for each extracted pixel block whose center of gravity is aligned to set a similarity between the extracted pixel blocks, and (iii) that generates representative images from the extracted pixel blocks based on the similarity; and

a vector converting unit that converts the representative images generated by the representative-image generating unit into vector information.

2. The image processing apparatus according to claim 1, further comprising:

a character recognizing unit that performs character recognition for the representative images generated by the representative-image generating unit; and

an associating unit that associates each result of the character recognition by the character recognizing unit with the corresponding vector information generated by the vector converting unit.

3. The image processing apparatus according to claim 1, further comprising:

a file generating unit that generates a file based on the vector information generated by the vector converting unit.

4. The image processing apparatus according to claim 1, further comprising:

an enlarging process unit that performs an enlarging process for each representative image generated by the representative-image generating unit, wherein

the vector converting unit converts, into vector information, each representative image for which the enlarging process unit has performed the enlarging process.

5. The image processing apparatus according to claim 4, wherein the enlarging process unit performs the enlarging process for each representative image using the plural pixel blocks, which are used by the representative-image generating unit to generate each representative image, so as to generate an enlarged representative image having a resolution higher than the used pixel blocks.

6. The image processing apparatus according to claim 5, wherein the enlarging process unit moves phases of the plural pixel blocks so that centers of gravity of the plural pixel blocks coincide with each other and interpolates pixels values of each enlarged representative image, so as to generate each enlarged representative image having the higher resolution.

7. The image processing apparatus according to claim 3, wherein

the extracting unit extracts a position of each extracted pixel block and assigns a second code to each extracted pixel block, and

the representative-image generating unit generates each representative image by bringing similar pixel blocks of the extracted pixel blocks together based on the similarity between the extracted pixel blocks,

the image processing apparatus further comprising:

a first-code assigning unit that assigns a first code to each representative image generated by the representative-image generating unit and associates each first code with the second codes of the pixel blocks corresponding to each first code;

an association-information generating unit that generates association information that associates each representative image generated by the representative-image generating unit with the corresponding first code assigned by the first-code assigning unit;

a character recognizing unit that performs character recognition for the pixel blocks extracted by the extracting unit; and

a code associating unit that associates each of third codes that are results of the character recognition performed by the character recognition unit with the corresponding first code assigned by the first-code assigning unit, wherein

the file generating unit that generates the file based on (i) the association information generated by the association-information generating unit, (ii) the first codes assigned by the first-code assigning unit, (iii) the positions of the pixel blocks corresponding to the first codes, the position being extracted by the extracting unit, and (iv) the association between the first codes and the third codes by the code associating unit.

8. The image processing apparatus according to claim 3, wherein

the extracting unit extracts a position of each extracted pixel block and assigns a second code to each extracted pixel block, and

the representative-image generating unit generates each representative image by bringing similar pixel blocks of the extracted pixel blocks together based on the similarity between the extracted pixel blocks,

the image processing apparatus further comprising:

a first-code assigning unit that assigns a first code to each representative image generated by the representative-image generating unit and associates each first code with the second codes of the pixel blocks corresponding to each first code;

an association-information generating unit that generates association information that associates each representative image generated by the representative-image generating unit with the corresponding first code assigned by the first-code assigning unit;

a character recognizing unit that performs character recognition for the pixel blocks extracted by the extracting unit; and

a code associating unit that associates each of third codes that are results of the character recognition performed by the character recognition unit with the corresponding first code assigned by the first-code assigning unit, wherein

the file generating unit generates the file based on (i) the association information generated by the association-information generating unit, (ii) the association between the first codes and the second codes by the first-code assigning unit, (iii) the positions of the pixel blocks corresponding to the second codes, and (iv) the association between the first codes and the third codes by the code associating unit.

9. The image processing apparatus according to claim 7, wherein

the character recognizing unit associates each second code of the pixel block for which the character recognition is performed with the corresponding third code, which is the result of the character recognition, and

the file generating unit changes (i) the association between the first codes and the second codes and (ii) the association between the first codes and the third codes based on the second codes and the third codes, which are associated by the character recognition unit, to generates the file.

10. The image processing apparatus according to claim 3, comprising:

the extracting unit extracts a position of each extracted pixel block and assigns a second code to each extracted pixel block, and

the representative-image generating unit generates each representative image by bringing similar pixel blocks of the extracted pixel blocks together based on the similarity between the extracted pixel blocks,

the image processing apparatus further comprising;

a first-code assigning unit that assigns a first code to each representative image generated by the representative-image generating unit and associates each first code with the second codes of the pixel blocks corresponding to each first code;

an association-information generating unit that generates association information that associates each representative image generated by the representative-image generating unit with the corresponding first code assigned by the first-code assigning unit;

a character recognizing unit that performs character recognition for each representative image generated by the representative-image generating unit and associates the second codes corresponding to each representative image with a third code that is a result of the character recognition; and

a code associating unit that associates each third code, which is the result of the character recognition performed by the character recognition unit, with the corresponding first code assigned by the first-code assigning unit, wherein

the file generating unit generates the file based on (i) the association information generated by the association-information generating unit, (ii) the association between the first codes and the second codes by the first-code assigning unit, (iii) the positions of the pixel blocks corresponding to the second codes, and (iv) the association between the first codes and the third codes by the code associating unit.

21. The image processing apparatus according to claim 1, wherein

the extracting unit extracts a position of each extracted pixel block;

the representative-image generating unit comprises

a similar-block extracting section extracts groups of pixel blocks from the pixel blocks extracted by the extracting unit, the pixel blocks belonging to each group being similar to each other, the similar-block extracting section that associates a code that uniquely identifies each group of the pixel blocks with each group of the pixel blocks,

a phase estimating section that estimates a phase of each pixel block based on the pixel blocks of the extracted groups, and

a high-resolution representative-image generating section that generates the representative image having a resolution higher than the pixel blocks of each group based on (i) the pixel blocks of each group extracted by the similar-block extracting section and (ii) the phases estimated by the phase estimating section, and

the vector converting unit converts the representative images generated by the high-resolution representative-image generating section into vector information,

the image processing apparatus further comprising;

an encoding unit that generates output data based on (i) the positions of the pixel blocks extracted by the extracting unit, (ii) the codes associated with the pixel blocks by the similar-block extracting section, (iii) the phases estimated by the phase estimating section, and (iv) the vector information generated by the vector converting unit.

22. An image processing apparatus comprising:

a decoding unit that decodes positions of pixel blocks, phases of the pixel blocks, a code of each representative image corresponding to the pixel blocks, and vector information of each representative image from data encoded by the image processing apparatus according to claim 21;

a representative-image generating unit that generates each representative image from the vector information decoded by the decoding unit;

an image storing unit that stores each representative image generated by the representative-image generating unit and the corresponding code decoded by the decoding unit in association with each other;

a low-resolution image generating unit that generates pixel blocks having a resolution lower than each representative image based on each representative image stored in the image storing unit and the phases decoded by the decoding unit; and

an image generating unit that generates an image based on the pixel blocks generated by the low-resolution image generating unit and the positions of the pixel blocks decoded by the decoding unit.

23. An image processing apparatus comprising:

a decoding unit that decodes positions of pixel blocks, phases of the pixel blocks, a code of each representative image corresponding to the pixel blocks, and vector information of each representative image from data encoded by the image processing apparatus according to claim 21;

a vector information storing unit that stores each vector information decoded by the decoding unit and the corresponding code decoded by the decoding unit in association with each other;

a representative-image generating unit that generates a representative image based on each vector information stored in the vector information storing unit and the phases decoded by the decoding unit; and

an image generating unit that generates an image based on the representative images generated by the representative-image generating unit and the positions of the pixel blocks decoded by the decoding unit.

25. A non-transitory computer-readable medium storing a program causing a computer to execute image processing, the image processing comprising:

decoding positions of pixel blocks, phases of the pixel blocks, a code of each representative image corresponding to the pixel blocks, and vector information of each representative image from data encoded by the image processing apparatus according to claim 21;

generating each representative image from the decoded vector information;

storing each representative image generated and the corresponding code decoded in association with each other;

generating pixel blocks having a resolution lower than each representative image based on each representative image stored and the phases decoded; and

generating an image based on the pixel blocks generated and the positions of the pixel blocks decoded.

26. A non-transitory computer-readable medium storing a program causing a computer to execute image processing, the image processing comprising:

decoding positions of pixel blocks, phases of the pixel blocks, a code of each representative image corresponding to the pixel blocks, and vector information of each representative image from data encoded by the image processing apparatus according to claim 21;

storing each vector information decoded and the corresponding code decoded in association with each other;

generating a representative image based on each vector information stored and the phases decoded; and

generating an image based on the representative images generated and the positions of the pixel blocks decoded.

11. A non-transitory computer-readable medium storing a program for causing a computer to execute image processing, the image processing comprising:

extracting pixel blocks from image data;

aligning each center of gravity of each extracted pixel block;

,

calculating a binary value for each extracted pixel block whose center of gravity is aligned to set a similarity between the extracted pixel blocks;

generating representative images from the extracted pixel blocks based on the similarity; and

converting the generated representative images into vector information.

12. The non-transitory computer-readable medium according to claim 11, wherein the image processing further comprises:

performing character recognition for the generated representative images; and

associating each result of the character recognition with the corresponding vector information.

13. The non-transitory computer-readable medium according to claim 11, wherein the image processing further comprises:

generating a file based on the generated vector information.

14. The non-transitory computer-readable medium according to claim 11, wherein the image processing further comprises:

performing an enlarging process for each generated representative image, wherein

the converting converts, into vector information, each representative image for which the enlarging process has been performed.

15. The non-transitory computer-readable medium according to claim 14, wherein the enlarging process is performed for each representative image using the plural pixel blocks, which are used to generate each representative image, so as to generate an enlarged representative image having a resolution higher than the used pixel blocks.

16. The non-transitory computer-readable medium according to claim 15, wherein the performing the enlarging process comprises moving phases of the plural pixel blocks so that centers of gravity of the plural pixel blocks coincide with each other and interpolates pixels values of each enlarged representative image, so as to generate each enlarged representative image having the higher resolution.

17. The non-transitory computer-readable medium according to claim 13, wherein

the extracting the pixel blocks comprises extracting a position of each extracted pixel block and assigning a second code to each extracted pixel block, and

the generating each representative image generates each representative image by bringing similar pixel blocks of the extracted pixel blocks together based on the similarity between the extracted pixel blocks,

the image processing further comprises:

assigning a first code to each generated representative image and associating each first code with the second codes of the pixel blocks corresponding to each first code;

generating association information that associates each generated representative image with the corresponding first code;

performing character recognition for the extracted pixel blocks; and

associating each of third codes that are results of the character recognition with the corresponding first code, wherein

the generating the file generates the file based on (i) the generated association information, (ii) the first codes, (iii) the extracted positions of the pixel blocks corresponding to the first codes, and (iv) the association between the first codes and the third codes.

18. The non-transitory computer-readable medium according to claim 13, wherein

the extracting the pixel blocks comprises extracting a position of each extracted pixel block and assigning a second code to each extracted pixel block, and

the generating each representative image generates each representative image by bringing similar pixel blocks of the extracted pixel blocks together based on the similarity between the extracted pixel blocks,

the image processing further comprises:

assigning a first code to each generated representative image and associating each first code with the second codes of the pixel blocks corresponding to each first code;

generating association information that associates each generated representative image with the corresponding first code;

performing character recognition for the extracted pixel blocks; and

associating each of third codes that are results of the character recognition with the corresponding first code, wherein

the generating the file generates the file based on (i) the generated association information, (ii) the association between the first codes and the second codes, (iii) the extracted positions of the pixel blocks, and (iv) the association between the first codes and the third codes.

19. The non-transitory computer-readable medium according to claim 17, wherein

the performing the character recognition comprises associating each second code of the pixel block for which the character recognition is performed with the corresponding third code, which is the result of the character recognition, and

the generating the file generates the file using the second codes and the third codes, which are associated with each other, with (i) the association between the first codes and the second codes and (ii) the association between the first codes and the third codes being reflected.

20. The non-transitory computer-readable medium according to claim 13, wherein

the extracting the pixel blocks comprises extracting a position of each extracted pixel block and assigning a second code to each extracted pixel block, and

the generating each representative image generates each representative image by bringing similar pixel blocks of the extracted pixel blocks together based on the similarity between the extracted pixel blocks,

the image processing further comprises:

assigning a first code to each generated representative image and associating each first code with the second codes of the pixel blocks corresponding to each first code;

generating association information that associates each generated representative image with the corresponding first code;

performing character recognition for each generated representative image and associating the second codes corresponding to each representative image with a third code that is a result of the character recognition; and

associating each third code, which is the result of the character recognition, with the corresponding first code, wherein

the generating the file generates the file based on (i) the generated association information, (ii) the association between the first codes and the second codes, (iii) the extracted positions of the pixel blocks corresponding to the second codes, and (iv) the association between the first codes and the third codes.

24. The non-transitory computer-readable medium according to claim 11, wherein

the extracting the pixel blocks comprises extracting a position of each extracted pixel block,

the generating the representative images comprises

extracting groups of pixel blocks from the extracted pixel blocks, the pixel blocks belonging to each group being similar to each other,

associating a code that uniquely identifies each group of the pixel blocks with each group of the pixel blocks,

estimating a phase of each pixel block based on the pixel blocks of the extracted groups, and

generating the representative image having a resolution higher than the pixel blocks of each group based on (i) the pixel blocks of each group and (ii) the estimated phases, and

the converting converts the generated representative images into vector information,

the image processing further comprising:

performing encoding based on (i) the positions of the extracted pixel blocks, (ii) the codes associated with the pixel blocks, (iii) the estimated phases, and (iv) the vector information.