Method and apparatus for identification of documents, and computer product

• US 20020018071A1
• Filed: 03/19/2001
• Published: 02/14/2002
• Est. Priority Date: 03/30/2000
• Status: Active Grant

• Current Assignee: Glory Limited
• Original Assignee: Glory Limited
• Inventors: Kameyama, Hirofumi, Ohnishi, Masatoshi
• Granted Patent: US 6,813,381 B2
• Time in Patent Office: Days
• US Class Current: 345/643
• CPC Class Codes: G06V 30/412 Layout analysis of documents structured with printed lines or input boxes, e.g. business forms or tables

Abstract

In the document identification apparatus, a ruled line feature extraction section determines a black pixel ratio of a document to be identified, and adds the black pixel ratio for each block to extract a ruled line feature. A ruled line feature verification section verifies the ruled line feature with a ruled line feature already registered in a ruled line feature dictionary to thereby identify the document. If identification is not possible with this procedure, a details judgment section verifies the image data in a specific area with the image data (characters or the like) registered in a specific area dictionary.

Claims

1. A document identification apparatus for discriminating various documents by comparing a feature quantity of image data of an input image of a document with a feature quantity of image data of at least one reference image stored beforehand, the document identification apparatus comprising:

a calculation unit which calculates a black pixel ratio, which black pixel ratio is a ratio of black pixels existing in a predetermined number of continuous pixels in horizontal or vertical direction from a specific pixel in the image data of the input image or the reference image; and

an extraction unit which divides the image data into a plurality of blocks, and separately adds the black pixel ratios corresponding to every pixel located in every block to extract a feature quantity of the image data.

2. The document identification apparatus according to claim 1, further comprising:

a memory which stores a feature quantity corresponding to the reference image extracted by the extraction unit; and

an identification unit which discriminates the document, when the feature quantity corresponding to the input image has been extracted by the extraction unit, by comparing a feature quantity corresponding to the input image with a feature quantity corresponding to the reference image stored in the memory.

3. The document identification apparatus according to claim 2, wherein the identification unit includes,

a candidate acquisition unit which acquires a plurality of document candidates according to a similarity between the feature quantity of the reference image stored in the memory and the feature quantity corresponding to the input image;

and

a specification unit which specifies a reference image corresponding to the input image, based on the image data of the reference image in each document candidate acquired by the candidate acquisition means and the image data of the input image.

4. A document identification method of discriminating various documents by comparing a feature quantity of image data of an input image of a document with a feature quantity of image data of at least one reference image stored beforehand, the document identification method comprising:

a calculation step of calculating a black pixel ratio, which black pixel ratio is a ratio of black pixels existing in a predetermined number of continuous pixels in horizontal or vertical direction from a specific pixel in the image data of the input image or the reference image; and

an extraction step of dividing the image data into a plurality of blocks, and separately adding the black pixel ratios corresponding to every pixel located in every block to extract a feature quantity of the image data.

5. The document identification method according to claim 4, further comprising:

a storage step of storing a feature quantity corresponding to the reference image extracted by the extraction step in a memory; and

an identification step of discriminating the document, when the feature quantity corresponding to the input image has been extracted in the extraction step, by comparing a feature quantity corresponding to the input image with a feature quantity corresponding to the reference image stored in the memory.

6. The document identification method according to claim 5, wherein the identification step includes,

a candidate acquisition step of acquiring a plurality of document candidates according to a similarity between the feature quantity of the reference image stored in the memory and the feature quantity corresponding to the input image;

and

a specification step of specifying a reference image corresponding to the input image, based on the image data of the reference image in each document candidate acquired in the candidate acquisition step and the image data of the input image.

7. A computer readable recording medium which stores a computer program which contains instructions which when executed on a computer realizes a document identification method of discriminating various documents by comparing a feature quantity of image data of an input image of a document with a feature quantity of image data of at least one reference image stored beforehand, the document identification method comprising:

a calculation step of calculating a black pixel ratio, which black pixel ratio is a ratio of black pixels existing in a predetermined number of continuous pixels in horizontal or vertical direction from a specific pixel in the image data of the input image or the reference image; and

an extraction step of dividing the image data into a plurality of blocks, and separately adding the black pixel ratios corresponding to every pixel located in every block to extract a feature quantity of the image data.

8. A computer readable recording medium according to claim 7, further comprising:

a storage step of storing a feature quantity corresponding to the reference image extracted by the extraction step in a memory; and

an identification step of discriminating the document, when the feature quantity corresponding to the input image has been extracted in the extraction step, by comparing a feature quantity corresponding to the input image with a feature quantity corresponding to the reference image stored in the memory.

9. A computer readable recording medium according to claim 7, wherein the identification step includes,

a candidate acquisition step of acquiring a plurality of document candidates according to a similarity between the feature quantity of the reference image stored in the memory and the feature quantity corresponding to the input image;

and

a specification step of specifying a reference image corresponding to the input image, based on the image data of the reference image in each document candidate acquired in the candidate acquisition step and the image data of the input image.

10. A computer program which contains instructions which when executed on a computer realizes a document identification method of discriminating various documents by comparing a feature quantity of image data of an input image of a document with a feature quantity of image data of at least one reference image stored beforehand, the document identification method comprising:

a calculation step of calculating a black pixel ratio, which black pixel ratio is a ratio of black pixels existing in a predetermined number of continuous pixels in horizontal or vertical direction from a specific pixel in the image data of the input image or the reference image; and

an extraction step of dividing the image data into a plurality of blocks, and separately adding the black pixel ratios corresponding to every pixel located in every block to extract a feature quantity of the image data.

11. A computer program according to claim 10, further comprising:

a storage step of storing a feature quantity corresponding to the reference image extracted by the extraction step in a memory; and

an identification step of discriminating the document, when the feature quantity corresponding to the input image has been extracted in the extraction step, by comparing a feature quantity corresponding to the input image with a feature quantity corresponding to the reference image stored in the memory.

12. A computer program according to claim 10, wherein the identification step includes,

a candidate acquisition step of acquiring a plurality of document candidates according to a similarity between the feature quantity of the reference image stored in the memory and the feature quantity corresponding to the input image;

and

a specification step of specifying a reference image corresponding to the input image, based on the image data of the reference image in each document candidate acquired in the candidate acquisition step and the image data of the input image.