Creating Flexible Structure Descriptions

• US 20130198615A1
• Filed: 07/31/2012
• Published: 08/01/2013
• Est. Priority Date: 08/01/2006
• Status: Active Grant

• Current Assignee: ABBYY Development LLC
• Original Assignee: ABBYY Software
• Inventors: Zuev, Konstantin, Tuganbaev, Diar, Filimonova, Irina, Zlobin, Sergey
• Granted Patent: US 8,908,969 B2
• Time in Patent Office: Days
• US Class Current: 715/243
• CPC Class Codes: G06F 40/106 Display of layout of documents; Previewing
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/1452 based on positionally close symbols, e.g. amount sign or URL-specific characters
• 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
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

In one embodiment, the invention provides a method, comprising detecting data fields on a scanned document image; generating a flexible document description based on the detected data fields, including creating a set of search elements for each data field, each search element having associated search criteria; and training or modifying the flexible document description using, for example, a search algorithm to detect the data fields on additional training images based on the set of search elements.

Claims

1. A method for creating a flexible structure description, the method comprising:

obtaining an image of a document;

detecting image objects on the image;

matching by a processor the image objects to a predetermined document type, wherein image objects distinguish the document type from other document types, and wherein image objects include anchor elements;

generating by the processor a flexible structure description corresponding to the predetermined document type based on the detected image objects, wherein the flexible structure description includes a set of search elements for each data field in the image of the document, each search element having an associated search criterion;

searching, via a search algorithm, additional document images to determine a respective document type of the additional document images, wherein each of the additional document images are of a document type corresponding to the predetermined document type;

modifying the flexible structure description based on said searching of additional document images, wherein the search algorithm is configured to detect data fields based on the flexible structure description, said data fields corresponding to the predetermined document type; and

repeating said searching and modifying of the flexible structure description until a defined level of precision is achieved or exceeded.

2. The method of claim 1, wherein said modifying of the flexible document description includes adding a type element to the set of search elements for a data field corresponding to a data type associated with the data field.

3. The method of claim 1, wherein detecting image objects on the image comprises:

detecting an anchor element on the image for a data field.

4. The method of claim 1, wherein the flexible document description includes a search algorithm that is capable of detecting a set of image objects corresponding to the predetermined document type.

5. The method of claim 1, wherein the method further comprises identifying a data type for each data field.

6. The method of claim 1, wherein the method further comprises:

detecting a string at a position relative to a data field and in a vicinity of the data field, wherein a text content of the string suggests an association between the data field and the string; and

modifying the flexible structure description to include a search element for the data field corresponding to the detected string.

7. The method of claim 1, wherein the method further comprises modifying the flexible structure description to include alternative search elements for a respective data field, and an instruction to search the alternative search elements consecutively when subsequently searching for the respective data field in a document image.

8. The method of claim 1, wherein the method further comprises:

before modifying the flexible structure description based on said searching of additional document images, for each particular additional document image that fails to correspond to a document type that matches the predetermined document type, displaying a prompt through a user interface indicating what data fields failed to be recognized corresponding to the predetermined document type; and

detecting an input corresponding to an identification of an anchor element associated with a data field associated with the predetermined document type.

9. A non-transitory computer-readable medium having stored thereon a sequence of instructions which when executed by a system, causes the system to perform a method, comprising:

obtaining an image of a document;

detecting image objects on the image;

matching by a processor the image objects to a predetermined document type, wherein image objects distinguish the document type from other document types, and wherein image objects include anchor elements;

generating by the processor a flexible structure description corresponding to the predetermined document type based on the detected image objects, wherein the flexible structure description includes a set of search elements for each data field in the image of the document, each search element having an associated search criterion;

searching, via a search algorithm, additional document images to determine a respective document type of the additional document images, wherein each of the additional document images are of a document type corresponding to the predetermined document type;

modifying the flexible structure description based on said searching of additional document images, wherein the search algorithm is configured to detect data fields based on the flexible structure description, said data fields corresponding to the predetermined document type; and

repeating said searching and modifying of the flexible structure description until a defined level of precision is achieved or exceeded.

10. The non-transitory computer-readable medium of claim 9, wherein said modifying of the flexible document description includes adding a type element to the set of search elements for a data field corresponding to a data type associated with the data field.

11. The non-transitory computer-readable medium of claim 9, wherein detecting image objects on the image comprises:

detecting an anchor element on the image for a data field.

12. The non-transitory computer-readable medium of claim 9, wherein the flexible document description includes a search algorithm that is capable of detecting a set of image objects corresponding to the predetermined document type.

13. The non-transitory computer-readable medium of claim 9, wherein the method further comprises identifying a data type for each data field.

14. The non-transitory computer-readable medium of claim 9, wherein the method further comprises:

detecting a string at a position relative to a data field and in a vicinity of the data field, wherein a text content of the string suggests an association between the data field and the string; and

modifying the flexible structure description to include a search element for the data field corresponding to the detected string.

15. The non-transitory computer-readable medium of claim 9, wherein the method further comprises modifying the flexible structure description to include alternative search elements for a respective data field, and an instruction to search the alternative search elements consecutively when subsequently searching for the respective data field in a document image.

16. The non-transitory computer-readable medium of claim 9, wherein the method further comprises:

before modifying the flexible structure description based on said searching of additional document images, for each particular additional document image that fails to correspond to a document type that matches the predetermined document type, displaying a prompt through a user interface indicating what data fields failed to be recognized corresponding to the predetermined document type; and

detecting an input corresponding to an identification of an anchor element associated with a data field associated with the predetermined document type.

17. A system comprising:

a processor; and

a memory coupled to the processor, the memory storing instructions which when executed by the processor causes the system to perform a method, the method comprising;

obtaining an image of a document;

detecting image objects on the image;

matching by a processor the image objects to a predetermined document type, wherein image objects distinguish the document type from other document types, and wherein image objects include anchor elements;

generating by the processor a flexible structure description corresponding to the predetermined document type based on the detected image objects, wherein the flexible structure description includes a set of search elements for each data field in the image of the document, each search element having an associated search criterion;

searching, via a search algorithm, additional document images to determine a respective document type of the additional document images, wherein each of the additional document images are of a document type corresponding to the predetermined document type;

modifying the flexible structure description based on said searching of additional document images, wherein the search algorithm is configured to detect data fields based on the flexible structure description, said data fields corresponding to the predetermined document type; and

repeating said searching and modifying of the flexible structure description until a defined level of precision is achieved or exceeded.

18. The system of claim 17, wherein said modifying of the flexible document description includes adding a type element to the set of search elements for a data field corresponding to a data type associated with the data field.

19. The system of claim 17, wherein detecting image objects on the image comprises:

detecting an anchor element on the image for a data field.

20. The system of claim 17, wherein the flexible document description includes a search algorithm that is capable of detecting a set of image objects corresponding to the predetermined document type.

21. The system of claim 17, wherein the method further comprises identifying a data type for each data field.

22. The system of claim 17, wherein the method further comprises:

detecting a string at a position relative to a data field and in a vicinity of the data field, wherein a text content of the string suggests an association between the data field and the string; and

modifying the flexible structure description to include a search element for the data field corresponding to the detected string.

23. The system of claim 17, wherein the method further comprises modifying the flexible structure description to include alternative search elements for a respective data field, and an instruction to search the alternative search elements consecutively when subsequently searching for the respective data field in a document image.

24. The system of claim 17, wherein the method further comprises:

before modifying the flexible structure description based on said searching of additional document images, for each particular additional document image that fails to correspond to a document type that matches the predetermined document type, displaying a prompt through a user interface indicating what data fields failed to be recognized corresponding to the predetermined document type; and

detecting an input corresponding to an identification of an anchor element associated with a data field associated with the predetermined document type.