Text character identification system and method thereof

• US 20050271275A1
• Filed: 06/01/2005
• Published: 12/08/2005
• Est. Priority Date: 06/04/2004
• Status: Active Grant

• Current Assignee: Yoshinaga Technologies LLC (Intellectual Ventures LLC)
• Original Assignee: BenQ Corporation (Qisda Corporation)
• Inventors: Chang, Hao-Wei
• Granted Patent: US 8,306,325 B2
• Time in Patent Office: Days
• US Class Current: 382/176
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/186 by deriving mathematical or geometrical properties from the whole image
• CPC Class Codes: G06V 30/413 Classification of content, e.g. text, photographs or tables

Abstract

A method for text character identification. The method acquires multiple connected components (CCs) in a binary image, and each CC has a pattern property value. The method determines at least one property limit based on the pattern property values, generates a filtering rule according to the property limit, and determines whether each of the CCs is a text character according to the filtering rule.

Claims

1. A method of text character identification, utilized to identify text characters in a binary image, comprising:

acquiring a plurality of connected components (CCs) in a binary image, each CC having a pattern property value;

determining at least one property limit based on the pattern property values;

generating a filtering rule according to the property limit; and

determining whether each of the CCs is a text character according to the filtering rule.

2. The method as claimed in claim 1 further comprising a step of converting a bitmap image to the binary image using a thresholding method.

3. The method as claimed in claim 2 wherein the thresholding method is minimum error thresholding method or Otsu thresholding method.

4. The method as claimed in claim 2 wherein in the step of acquiring the CCs, the CCs are acquired using a region growing method.

5. The method as claimed in claim 4 wherein the region growing method merges pixels to CCs when those pixels are both adjacent to the CCs and similar in property.

6. The method as claimed in claim 1 wherein the pattern property value is a CC height, CC width or CC density.

7. The method as claimed in claim 5 wherein the property limit comprises an upper height limit, an upper width limit, an upper density limit, a lower height limit, a lower width limit and a lower density limit, in the step of determining the property limit further comprises-the steps of:

acquiring a base height being the most frequently occurring in the binary image;

determining the upper and lower height limits by respectively multiplying/adding a first weighted value by/to the base height and dividing/subtracting the same or different weighted value into/from the base width;

acquiring a base width being the most frequently occurring in the binary image;

determining upper and lower width limits by respectively multiplying/adding a second weighted value by/to the base width and dividing/subtracting the same or different weighted value into/from the base width; and

acquiring upper and lower density limits.

8. The method as claimed in claim 7 wherein the filtering rule comprises:

one of the CCs is not a text character if its height is lower than the lower height limit and its width is lower than the lower width limit;

one of the CCs is not a text character if its height is higher than the upper height limit or its width is higher than the upper width limit; and

one of the CCs is not a text character if its density is higher than the upper density limit or its density is lower than the lower density limit.

9. The method as claimed in claim 1 further comprising the steps of:

merging neighboring CCs into a plurality of text regions; and

determining the CCs in any text region with the number of CCs being lower than a count threshold, as not being text characters.

10. A computer-readable storage medium storing a computer program providing a method of text character identification, the method comprising using a computer to perform the steps of:

acquiring a plurality of connected components (CCs) in a binary image, each CC having a pattern property value;

determining at least one property limit based on the pattern property values;

generating a filtering rule according to the property limit; and

determining whether each of the CCs is a text character according to the filtering rule.

11. A system of text character identification, utilized to identify text characters in a binary image, comprising:

a connected component acquisition unit configured to acquire a plurality of connected components (CCs) in a binary image, each CC having a pattern property value;

a limit generation unit configured to determine at least one property limit based on the pattern property values;

a filtering rule generation unit configured to generate a filtering rule according to the property limit; and

a text determination unit configured to determine whether each of the CCs is a text character according to the filtering rule.

12. The system as claimed in claim 11 further comprising a binary image generation unit configured to convert a bitmap image to the binary image using a thresholding method.

13. The system as claimed in claim 12 wherein the thresholding method is minimum error thresholding method or Otsu thresholding method.

14. The system as claimed in claim 11 wherein the CCs are acquired using a region growing method.

15. The system as claimed in claim 14 wherein the region growing method merges pixels to CCs when those pixels are both adjacent to the CCs and similar in property.

16. The system as claimed in claim 11 wherein the pattern property value is a CC height, CC width or CC density.

17. The system as claimed in claim 15 wherein the property limit comprises an upper height limit, an upper width limit, an upper density limit, a lower height limit, a lower width limit and a lower density limit.

18. The system as claimed in claim 17 wherein the limit generation module acquires a base height being the most frequently occurring in the binary image, determines the upper and lower height limits by respectively multiplying/adding a first weighted value by/to the base height and dividing/subtracting the same or different weighted value into/from the base width, acquires a base width being the most frequently occurring in the binary image, determines upper and lower width limits by respectively multiplying/adding a second weighted value by/to the base width and dividing/subtracting the same or different weighted value into/from the base width, and acquires upper and lower density limits.

19. The system as claimed in claim 18 wherein the filtering rule comprises:

one of the CCs is not a text character if its height is lower than the lower height limit and its width is lower than the lower width limit;

one of the CCs is not a text character if its height is higher than the upper height limit or its width is higher than the upper width limit; and

one of the CCs is not a text character if its density is higher than the upper density limit or its density is lower than the lower density limit.

20. The system as claimed in claim 11 wherein the text determination module merges neighboring CCs into a plurality of text regions, and determines the CCs in any text region wherein the number of CCs being lower than a count threshold, as not being text characters.