METHOD AND SYSTEM FOR TEXT SEGMENTATION

• US 20120281919A1
• Filed: 05/06/2011
• Published: 11/08/2012
• Est. Priority Date: 05/06/2011
• Status: Abandoned Application

• Current Assignee: King Abdul AZIZ City For Science and Technology
• Original Assignee: King Abdul AZIZ City For Science and Technology
• Inventors: Al-Omari, Hussein Khalid, Khorsheed, Mohammad Sulaiman, Abdulkader, Ahmad
• Application Number:
  US13/102,373
• Publication Number:
  US 20120281919A1
• Time in Patent Office: Days
• US Class Current: 382/177
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/148 Segmentation of character regions
• CPC Class Codes: G06V 30/244 using graphical properties, e.g. alphabet type or font
• CPC Class Codes: G06V 30/293 of characters other than Kanji, Hiragana or Katakana

Abstract

A method and system for segmenting a text into a plurality of sections is provided. The text may be received in the form of an image. The method involves receiving one or more input labels from a user corresponding to one or more segmentation points of a plurality of segmentation points of the text. The plurality of segmentation points of the text are obtained by applying one or more segmentation heuristics over the text. The one or more input labels provided by the user are utilized to label the plurality of segmentation points of the text. In response to labeling, validation is performed to identify whether a segmentation point of the plurality of segmentation points is a valid segmentation point. Thereafter, based on the validation, a set of valid segmentation points is updated with one or more segmentation points of the plurality of segmentation points. The set of valid segmentation points facilitates segmentation of the text for recognizing the plurality of sections.

Claims

1. A method of segmenting a text into a plurality of text segments, the method comprising:

receiving at least one input label from a user corresponding to at least one segmentation point of a plurality of segmentation points of the text;

analyzing the plurality of segmentation points based on the at least one input label received;

validating whether a segmentation point of the plurality of segmentation points is indicative of a valid segmentation point in response to analyzing; and

updating a set of valid segmentation points with at least one segmentation point of the plurality of segmentation points based on the validation, wherein the set of valid segmentation points facilitates segmentation of the text into the plurality of text segments.

2. The method of claim 1, wherein the text is an Arabic text.

3. The method of claim 1, wherein the text is in form of an image.

4. The method of claim 1, wherein a valid segmentation point separates at least two text segments of the plurality of text segments of the text.

5. The method of claim 1, wherein a text segment of the plurality of text segments comprises at least one of characters, words, and lines.

6. The method of claim 1, wherein the plurality of segmentation points are obtained by applying at least one segmentation heuristic to the text.

7. The method of claim 1, wherein an input label corresponds to a valid segmentation point of the text.

8. The method of claim 1, wherein an input label is an answer for a query, the query is generated corresponding to a segmentation point of the at least one segmentation point.

9. The method of claim 1, wherein an input label is a binary value, the binary value indicates validity of a segmentation point.

10. The method of claim 1, wherein analyzing comprises mapping a segmentation point corresponding to an input label with the plurality of segmentation points.

11. The method of claim 10, wherein analyzing further comprises marking a segmentation point of the plurality of segmentation points based on mapping with the segmentation point corresponding to the input label.

12. The method of claim 1, wherein validating comprises assigning a confidence value to a segmentation point corresponding to an input label, wherein the confidence value denotes a probability of the segmentation point indicative of a valid segmentation point.

13. The method of claim 12, wherein validating further comprises determining whether the confidence value is above or below a predefined threshold limit.

14. The method of claim 13, wherein updating comprises appending a segmentation point corresponding to an input label of the at least one input label in the set of valid segmentation points, wherein a confidence value of the segmentation point is above a predefined threshold limit.

15. A computer program product comprising a computer usable medium having a computer readable program method for segmenting a text into a plurality of text segments, wherein the computer readable program when executed on a computer causes the computer to:

receive at least one input label from a user corresponding to at least one segmentation point of a plurality of segmentation points of the text;

analyze the plurality of segmentation points based on the at least one input label received;

validate whether a segmentation point of the plurality of segmentation points is indicative of a valid segmentation point based on analyzing; and

update a set of valid segmentation points with a segmentation point of the plurality of segmentation points based on the validation, wherein the set of valid segmentation points facilitates segmentation of the text into the plurality of text segments.

16. The computer program product of claim 15, wherein the text is an Arabic text.

17. The computer program product of claim 15, wherein the plurality of segmentation points are obtained by applying at least one segmentation heuristic to the text.

18. The computer program product of claim 15 further comprising computer readable program method that causes the computer to analyze by:

mapping a segmentation point corresponding to an input label with the plurality of segmentation points; and

marking a segmentation point of the plurality of segmentation points based on mapping with the segmentation point corresponding to the input label.

19. The computer program product of claim 15 further comprising computer readable program method that causes the computer to validate by:

assigning a confidence value to a segmentation point corresponding to an input label of the at least one input label, wherein the confidence value is probability of the segmentation point indicative of a valid segmentation point; and

determining whether the confidence value is above or below a threshold limit.

20. The computer program product of claim 19 further comprising computer readable program method that causes the computer to update by appending a segmentation point corresponding to an input label of the at least one input label in the set of valid segmentation points, wherein a confidence value of the segmentation point is above a predetermined threshold limit.

21. A system for segmenting a text into a plurality of text segments, the system comprising,

a processor, wherein the processor is configured to:

receive at least one input label from a user corresponding to at least one segmentation point of a plurality of segmentation points of the text;

analyze the plurality of segmentation points based on the at least one input label received;

validate whether a segmentation point of the plurality of segmentation points is indicative of a valid segmentation point in response to labeling; and

update a set of valid segmentation points with at least one segmentation point of the plurality of segmentation points based on the validation, wherein the set of valid segmentation points facilitates segmentation of the text into the plurality of text segments; and

a memory for storing the set of valid segmentation points.

22. The system of claim 21, wherein the text is an Arabic text.

23. The system of claim 21, wherein the processor is configured to analyze by:

mapping a segmentation point corresponding to an input label with the plurality of segmentation points; and

marking a segmentation point of the plurality of segmentation points based on mapping with the segmentation point corresponding to the input label.

24. The system of claim 21, wherein the processor is configured to validate by:

assigning a confidence value to a segmentation point corresponding to an input label, wherein the confidence value denotes a probability of the segmentation point indicative of a valid segmentation point; and

determining whether the confidence value is above or below a predefined threshold limit.

25. The system of claim 24, wherein the processor is configured to update the set of valid segmentation points by appending a segmentation point corresponding to an input label of the at least one input label in the set of valid segmentation points, wherein a confidence value of the segmentation point is above a predefined threshold limit.