Claims:WE CLAIM
1. A method of determining structural blocks of a document, the method comprising:
extracting, by a document analysis device, a plurality of text lines from the document;
generating, by the document analysis device, a feature vector for each of the plurality of text lines by determining a set of feature values for a set of corresponding features in each of the plurality of text lines;
determining, by the document analysis device, at least one dominant feature from among the set of corresponding features and at least one corresponding dominance factor, for each of a plurality of structural classes, based on the feature vector for each of the plurality of text lines;
deriving, by the document analysis device, a set of rules for classification of the plurality of text lines into the plurality of structural classes, wherein each of the set of rules correspond to one of the plurality of structural classes and is based on the at least one dominant feature and the at least one corresponding dominance factor for that class; and
determining, by the document analysis device, a structural block tag for each of the plurality of text lines based on the set of rules.

2. The method of claim 1, further comprising:
receiving an image document;
performing an optical text recognition on the image document to generate the document.

3. The method of claim 1, wherein extracting the plurality of text lines comprises applying a text extraction tool with a pre-defined or a dynamic threshold on the document.

4. The method of claim 1, wherein the set of corresponding features comprise at least one of a positional feature, a font feature, a count feature, or a spacing feature.

5. The method of claim 1, wherein the set of feature values for the set of corresponding features comprise at least one of positional coordinates of a text line, a font size in the text line, one or more flags for one or more font styles, a length of the text line, or a spacing between at least two of the plurality of text line.

6. The method of claim 1, wherein determining the at least one dominant feature comprises comparing the feature vector for a text line with a set of feature vectors of a set of neighboring text lines, and wherein the set of neighboring text lines comprises at least one of a pre-defined set of preceding text lines, or a pre-defined set of successive text lines.

7. The method of claim 1, wherein each of the set of rules comprises a sum of the at least one dominant feature modified by the at least one corresponding dominance factor.

8. The method of claim 1, further comprising determining at least one corresponding threshold value for the at least one dominant feature, and wherein each of the set of rules comprises a sum of the at least one dominant feature modified by the at least one corresponding dominance factor and the at least one corresponding threshold value.

9. The method of claim 8, wherein determining the at least one corresponding threshold value for the at least one dominant feature comprises:
initializing the at least one corresponding threshold value based on an initial difference between feature values for the at least one dominant feature; and
dynamically adjusting the at least one corresponding threshold value in each of a plurality of subsequent passes.

10. The method of claim 1, wherein the structural block tag comprises one of a paragraph tag, a paragraph start tag, a paragraph end tag, a single line tag, a title tag, a section header tag, a footnote tag, a list tag, or a table of content tag.

11. The method of claim 1, wherein determining the structural block tag comprises:
determining a set of scores for a text line based on the set of rules; and
determining the structural block tag corresponding to one of the plurality of structural classes based on an indicative score among the set of scores.

12. A system for determining structural blocks of a document, the system comprising:
a document analysis device comprising at least one processor and a computer-readable medium storing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
extracting a plurality of text lines from the document;
generating a feature vector for each of the plurality of text lines by determining a set of feature values for a set of corresponding features in each of the plurality of text lines;
determining at least one dominant feature from among the set of corresponding features and at least one corresponding dominance factor, for each of a plurality of structural classes, based on the feature vector for each of the plurality of text lines;
deriving a set of rules for classification of the plurality of text lines into the plurality of structural classes, wherein each of the set of rules correspond to one of the plurality of structural classes and is based on the at least one dominant feature and the at least one corresponding dominance factor for that class; and
determining a structural block tag for each of the plurality of text lines based on the set of rules.

13. The system of claim 12, wherein extracting the plurality of text lines comprises applying a text extraction tool with a pre-defined or a dynamic threshold on the document.

14. The system of claim 12, wherein the set of corresponding features comprise at least one of a positional feature, a font feature, a count feature, or a spacing feature, and wherein the set of feature values for the set of corresponding features comprise at least one of positional coordinates of a text line, a font size in the text line, one or more flags for one or more font styles, a length of the text line, or a spacing between at least two of the plurality of text line.

15. The system of claim 12, wherein determining the at least one dominant feature comprises comparing the feature vector for a text line with a set of feature vectors of a set of neighboring text lines, and wherein the set of neighboring text lines comprises at least one of a pre-defined set of preceding text lines, or a pre-defined set of successive text lines.

16. The system of claim 12, wherein each of the set of rules comprises a sum of the at least one dominant feature modified by the at least one corresponding dominance factor.

17. The system of claim 12, wherein the operations further comprise determining at least one corresponding threshold value for the at least one dominant feature, and wherein each of the set of rules comprises a sum of the at least one dominant feature modified by the at least one corresponding dominance factor and the at least one corresponding threshold value.

18. The system of claim 17, wherein determining the at least one corresponding threshold value for the at least one dominant feature comprises:
initializing the at least one corresponding threshold value based on an initial difference between feature values for the at least one dominant feature; and
dynamically adjusting the at least one corresponding threshold value in each of a plurality of subsequent passes.

19. The system of claim 12, wherein determining the structural block tag comprises:
determining a set of scores for a text line based on the set of rules; and
determining the structural block tag corresponding to one of the plurality of structural classes based on an indicative score among the set of scores.

Dated this 30th day of July, 2018

Swetha SN
Of K&S Partners
Agent for the Applicant
IN/PA-2123
, Description:TECHNICAL FIELD
This disclosure relates generally to document processing, and more particularly to method and system for determining structural blocks of a document.