AUTOMATIC FILE NAME GENERATION IN OCR SYSTEMS

• US 20100215272A1
• Filed: 03/30/2010
• Published: 08/26/2010
• Est. Priority Date: 09/23/2008
• Status: Active Grant

• Current Assignee: ABBYY Production LLC (ABBYY Software)
• Original Assignee: ABBYY Development LLC
• Inventors: ISAEV, ANDREY, Deryagin, Dmitry, Anisimovich, Konstantin
• Granted Patent: US 8,452,132 B2
• Time in Patent Office: Days
• US Class Current: 382/182
• 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

Methods and system for processing document images in OCR systems, particularly for selecting a proper file name for a recognized document. The method comprises generating at least one document type hypothesis for the document; verifying each document type hypothesis; selecting a best document type hypothesis and saving the document with a proper name based on the best type hypothesis and unique features. The method further includes determining a logical structure of a document and selecting a best document model hypothesis that has the best degree of correspondence with the selected best block hypotheses for the document. On the basis of the best document model hypothesis the text document reflecting the logical structure of the source document in extended computer-editable format is formed and saved with a proper file name.

Claims

1-20. -20. (canceled)

21. A method for processing a document comprising:

analyzing the document to determine a type of the document comprising

generating at least one document type hypothesis;

verifying the at least one document type hypothesis by searching for features, keywords, and structural elements that are distinctive for the type of the document and by searching for features which are unique for the document;

selecting a best document type hypothesis from the at least one document type hypothesis; and

saving in a memory device the document with a mine based on the best document type hypothesis and the found unique features.

22. The method of claim 21, wherein the analyzing comprises analyzing the document using Optical Character Recognition techniques.

23. The method of claim 21, wherein analyzing the document further comprises determining a logical structure of the document comprising:

generating at least one document model hypothesis for the document;

verifying the at least one document model hypothesis by

generating at least one block hypothesis for each block in the document by searching for keywords, structural elements, and unique features that are distinctive for the block hypothesis; and

selecting a best block hypothesis for each block from the at least one block hypothesis;

selecting a best document model hypothesis having a best degree of correspondence with the selected best block hypothesis for each block in the document.

24. The method of claim 23, wherein determining the logical structure of the document comprises performing physical structure analysis to identify each block of the document.

25. The method of claim 23, wherein the name of the saved document is based on at least one of the best document type hypothesis and the best document model hypothesis and found unique features.

26. The method of claim 23, wherein saving the document in a memory device, as a prior step, further comprises forming the document in extended computer-editable format, wherein forming the document is based on the best document model hypothesis.

27. The method of claim 23, wherein the generating document model hypothesis is based on information about mutual arrangement of blocks, keywords, structural elements, and unique features that comprise the document.

28. A system for processing a document comprising:

at least one memory; and

at least a processor controlled by stored programmed instructions in the at least one memory to perform functions of

analyzing the document to determine a type of the document comprising

generating at least one document type hypothesis;

verifying the at least one document type hypothesis by searching for features, keywords, and structural elements that are distinctive for the type of the document and by searching for features which are unique for the document;

selecting a best document type hypothesis from the at least one document type hypothesis; and

saving in a memory device the document with a name based on the best document type hypothesis and the found unique features.

29. The system of claim 28, wherein the analyzing comprises analyzing the document in a computer-editable format using Optical Character Recognition techniques.

30. The system of claim 28, wherein analyzing the document further comprises determining a logical structure of the document comprising:

generating at least one document model hypothesis for the document;

verifying the at least one document model hypothesis by

generating at least one block hypothesis for each block in the document by searching for keywords, structural elements, and unique features that are distinctive for the block hypothesis; and

selecting a best block hypothesis for each block from the at least one block hypothesis; and

selecting a best document model hypothesis having a best degree of correspondence with the selected best block hypothesis for each block in the document.

31. The system of claim 30, wherein determining the logical structure of the document comprises performing physical structure analysis to identify each block of the document.

32. The system of claim 30, wherein the name of the saved document is based on at least one of the best document type hypothesis and the best document model hypothesis and found unique features.

33. The system of claim 30, wherein the generating document model hypothesis is based on information about mutual arrangement of blocks, keywords, structural elements, and unique features that comprise the document.

34. The system of claim 28, wherein the generating document type hypothesis is based on results of preliminary training of the processor on a sufficient set of documents of various types.

35. The system of claim 30, wherein the generating document model hypothesis is based on results of preliminary training of the processor on a sufficient set of documents of various types.

36. A computer readable medium containing a computer program product for processing a document, the computer program product comprising:

program code for analyzing the document to determine a type of the document comprising

program code for generating at least one document type hypothesis;

program code for verifying the at least one document type hypothesis by searching for features, keywords, and structural elements that are distinctive for the type of the document and by searching for features which are unique for the document;

program code for selecting a best document type hypothesis from the at least one document type hypothesis; and

program code for saving in a memory device the document with a name based on the best document type hypothesis and the found unique features.

37. The computer program product of claim 36, wherein the program code for analyzing comprises program code for analyzing the document in a computer-editable format using Optical Character Recognition techniques.

38. The computer program product of claim 36, wherein the program code for analyzing the document further comprises program code for determining a logical structure of the document comprising:

program code for generating at least one document model hypothesis for the document;

program code for verifying the at least one document model hypothesis by

program code for generating at least one block hypothesis for each block in the document by searching for keywords, structural elements, and unique features that are distinctive for the block hypothesis; and

program code for selecting a best block hypothesis for each block from the at least one block hypothesis; and

program code for selecting a best document model hypothesis having a best degree of correspondence with the selected best block hypothesis for each block in the document.

39. The computer program product of claim 38, wherein program code for determining the logical structure of the document comprises program code for performing physical structure analysis to identify each block of the document.

40. The computer program product of claim 38, wherein the name of the saved document is based on at least one of the best document type hypothesis and the best document model hypothesis and found unique features.

41. The computer program product of claim 38, wherein the generating document model hypothesis, is based on information about mutual arrangement of blocks, keywords, structural elements, and unique features that comprise the document.

42. The method of claim 38, wherein saving the document in a memory device, as a prior step, further comprises forming the document in extended computer-editable format, wherein forming the document is based on the best document model hypothesis.