IDENTIFYING PICTURE AREAS BASED ON GRADIENT IMAGE ANALYSIS

• US 20110013847A1
• Filed: 07/16/2010
• Published: 01/20/2011
• Est. Priority Date: 07/18/2009
• Status: Active Grant

• Current Assignee: ABBYY Production LLC (ABBYY Software)
• Original Assignee: ABBYY Software
• Inventors: DERYAGIN, DMITRY, STATSENKO, KONSTANTIN
• Granted Patent: US 9,092,668 B2
• Time in Patent Office: Days
• US Class Current: 382/224
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/18076 by analysing connectivity, e.g. edge linking, connected component analysis or slices
• CPC Class Codes: G06V 30/18162 related to a structural representation of the pattern
• CPC Class Codes: G06V 30/413 Classification of content, e.g. text, photographs or tables

Abstract

In one embodiment, a method for identifying areas in a document image is provided. The method comprises generating binarized and gradient images based on the document image; and performing a classification operation to classify areas in the document image into one of a noise area and a picture area based on attributes computed on the binarized and gradient images.

Claims

1. A method for identifying areas in a document image, the method comprising:

generating binarized and gradient images based on the document image; and

performing a classification operation to classify each area in the document image into a class based on attributes computed on the binarized and gradient images;

wherein the class comprises at least one of a picture area, a text area and a noise area.

2. The method of claim 1, wherein the gradient images comprise a vertical gradient image, a horizontal gradient image, and a combined horizontal and vertical gradient image.

3. The method of claim 1, further comprising performing a preliminary segmentation operation to segment the document image into preliminary areas prior to performing the classification operation on the preliminary areas.

4. The method of claim 1, wherein the classification operation uses a plurality of classifiers.

5. The method of claim 1, wherein the attributes comprise an average length of black and white vertical and horizontal Run-Length Encoding (RLE) strokes computed on the binarized image.

6. The method of claim 1, wherein the attributes are characterized in terms of average value, dispersion, asymmetry coefficient, and a coefficient of excess computed on the horizontal and vertical gradient images.

7. The method of claim 1, wherein the attributes comprises an average number of white holes in connected black components of the combined gradient images.

8. The method of claim 1, wherein the classification operation comprises combining the attributes and detecting the class based on the combined attributes.

9. The method of claim 8, wherein detecting the class comprises computing a probability density for each possible class, comparing the values of the probability densities, and attributing the area to the class with the greater probability density.

10. The method of claim 8, wherein detecting the class comprises computing a probability density for each possible class, comparing the value of the probability density with an empirically determined constant, and attributing the area to the selected class, if its probability density is greater than the constant.

12. The document analysis system of claim 10, wherein the gradient images comprise a vertical gradient image, a horizontal gradient image, and a combined horizontal and vertical gradient image.

13. The document analysis system of claim 10, wherein the method further comprises performing a preliminary segmentation operation to segment the document image into preliminary areas prior to performing the classification operation on the preliminary areas.

14. The document analysis system of claim 10, wherein the classification operation uses a plurality of classifiers.

11. A document analysis system, comprising:

a processor; and

a memory coupled to the processor;

the memory storing instructions to perform a document analysis method for identifying areas in a document image, the method comprising;

generating binarized and gradient images based on the document image; and

performing a classification operation to classify each area in the document image into a class based on attributes computed on the binarized and gradient images;

wherein the class comprises at least one of a picture area, a text area and a noise area.

15. The document analysis system of claim 11, wherein the attributes comprise an average length of black and white vertical and horizontal Run-Length Encoding (RLE) strokes computed on the binarized image.

16. The document analysis system of claim 11, wherein the attributes are characterized in terms of average value, dispersion, asymmetry coefficient, and a coefficient of excess computed on the horizontal and vertical gradient images.

17. The document analysis system of claim 11, wherein the attributes comprises an average number of white holes in connected black components of the combined gradient images.

18. A computer-readable medium having stored thereon a sequence of instructions which when executed by a processing system cause the system to perform a method for identifying areas in a document image, the method comprising:

generating binarized and gradient images based on the document image; and

performing a classification operation to classify each area in the document image into a class based on attributes computed on the binarized and gradient images;

wherein the class comprises at least one of a picture area, a text area and a noise area.

19. The computer-readable medium of claim 18, wherein the gradient images comprise a vertical gradient image, a horizontal gradient image, and a combined horizontal and vertical gradient image.

20. The computer-readable medium of claim 18, wherein the method further comprises performing a preliminary segmentation operation to segment the document image into preliminary areas prior to performing the classification operation on the preliminary areas.

21. The computer-readable medium of claim 18, wherein the classification operation uses a plurality of classifiers.

22. The computer-readable medium of claim 18, wherein the attributes comprise an average length of black and white vertical and horizontal Run-Length Encoding (RLE) strokes computed on the binarized image.

23. The computer-readable medium of claim 18, wherein the attributes are characterized in terms of average value, dispersion, asymmetry coefficient, and a coefficient of excess computed on the horizontal and vertical gradient images.

24. The computer-readable medium of claim 18, wherein the attributes comprises an average number of white holes in connected black components of the combined gradient images.