PAGE CLASSIFIER ENGINE

• US 20090144605A1
• Filed: 12/03/2007
• Published: 06/04/2009
• Est. Priority Date: 12/03/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: DRESEVIC, BODIN, RADAKOVIC, BOGDAN, UZELAC, ALEKSANDAR, TRUTNER, OREN
• Granted Patent: US 8,392,816 B2
• Time in Patent Office: Days
• US Class Current: 715/200
• CPC Class Codes: G06V 30/416 Extracting the logical structure, e.g. chapters, sections or page numbers; Identifying elements of the document, e.g. authors

Abstract

Embodiments of the present invention relate to classifying pages of an electronic document, such as a scanned book page. OCR software is applied to the contents of the electronic document, revealing semantic information about the content of the electronic document. Software-based features are applied to the semantic information to determine the type of page the electronic document is. Page types may include table of contents (TOC), table of figures (TOF), bibliography, index, or other types of pages commonly found in a book, magazine, or other publication. Once determined, the determined page type is stored and used by other software engines.

Claims

1. One or more computer-storage media having computer-executable instructions embodied thereon that, when executed, perform a method for classifying a page type of a portion of an electronic document, the method comprising:

receiving an OCR file associated with the portion of the electronic document, wherein the OCR file includes semantic information about text in the portion of the electronic document;

applying one or more features to the semantic information;

based on the application of the one or more features to the semantic information, determining the page type of the portion of the electronic document; and

storing an indication of the page type.

2. The media of claim 1, wherein the OCR file includes metadata indicating an adjusted page number of the electronic document.

3. The media of claim 1, wherein the semantic information includes one or more indications of the bounding boxes of each line in the portion of the electronic document.

4. The media of claim 3, wherein the semantic information includes one or more indications of the bounding boxes of each word in each line of the portion of the electronic document.

5. The media of claim 1, wherein the semantic information includes an indication of the underlying meaning of one or more words in the portion of the electronic document.

6. The media of claim 1, wherein the one or more features comprise software-implemented rules for incrementing a score based on characteristics of the semantic information.

7. The media of claim 1, wherein the one or more features comprise one or more software-implemented rules for determining whether text within the portion of the electronic document varies in one of at least size, font, or format from other text in the portion of the electronic document.

8. The media of claim 1, wherein the one or more features comprise searching the text on a page number listed in the portion of the electronic document to determine whether text on the portion of the electronic document is contained on the page corresponding to the page number.

9. The media of claim 1, wherein the one or more features comprise:

incrementing a score based on the occurrence of an event; and

determining whether the score is above a threshold value; and

assigning a page type to the portion of the electronic document if the score exceeds the threshold value.

10. The media of claim 1, further comprising:

determining a page type of the page preceding the electronic document; and

in part, basing the determination of the page type on the page type of the page preceding the electronic document.

11. The media of claim 1, wherein the one or more features comprise determining whether page numbers after a separator in the portion of the electronic document are in succession.

12. The media of claim 1, wherein storing an indication of the page type further comprises manipulating metadata associated with the portion of the electronic document.

13. The media of claim 1, wherein storing an indication of the page type further comprises appending an OCR file.

14. A method for determining a page type of a portion of an electronic document, comprising:

receiving an OCR file associated with the portion of the electronic document, wherein the OCR file includes semantic information about text in the portion of the electronic document;

analyzing a portion of the semantic information by applying one or more features to the semantic information;

based on the application of the one or more features, determining the page type of the portion of the electronic document; and

storing an indication of the page type.

15. The computer system of claim 14, wherein the one or more features are software-implemented rules that have been determined by a machine learning network.

16. The computer system of claim 15, wherein the machine learning network includes at least one neural network.

17. The computer system of claim 14, wherein the machine learning network utilizes a Bayesian algorithm.

18. The computer system of claim 17, wherein the page type is indicative of at least one of a table of contents, table of figures, index, bibliography, foreword, epilogue, or prologue page.

19. A computerized method for storing an indication that an electronic document is a table of contents (TOC) page type, comprising:

receiving semantic information extracted from the electronic document, the semantic information including an indication of the underlying meaning of at least one word;

applying one or more features to the semantic information;

determining that the electronic document is a TOC page type based on the underlying meaning of the at least one word; and

storing an indication indicating that the electronic document is a TOC page type.

20. The method of claim 19, wherein the electronic document comprises a scanned textual image.