Creating flexible structure descriptions of documents with repetitive non-regular structures

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

receiving an image of a document of a particular document type that contains a table;

receiving an entry describing an item in the table;

searching for title elements based upon the entry;

detecting data fields and anchor elements for the entry;

generating, using a processor, a flexible structure description for the particular document type that includes a set of search elements for each data field in the image of the document and the title elements;

matching the flexible structure description against the image; and

extracting data from the image based upon the matching of the flexible structure description against the image.

2. The method of claim 1 , further comprising adjusting, using the processor, the flexible structure description based on user corrections of detected data fields, title elements, and/or anchor elements.

3. The method of claim 1 , wherein the table spans multiple pages of the document, and wherein the title elements repeat on two or more pages of the multiple pages.

4. The method of claim 1 , wherein the generating the flexible structure description comprises identifying anchor elements for searching data fields.

5. The method of claim 1 , wherein the generating the flexible structure description comprises generating values of search parameters for each element and field.

6. The method of claim 1 , wherein the matching the flexible structure description against the image comprises:

matching the title elements on each page of the document;

removing the matched title elements from a search area of the document; and

searching the search area with the matched title elements removed for data fields.

7. The method of claim 1 , wherein the entry corresponds to a row of the table that spans multiple lines of the document.

8. The method of claim 1 , further comprising receiving at least one entry, wherein the entry comprises multiple cells of the table, and wherein at least two of the multiple cells overlap.

9. A system for creating a flexible structure description, the system comprising:

one or more electronic processors configured to:

receive an image of a document of a particular document type that contains a table;

receive an entry describing an item in the table;

search for title elements based upon the entry;

detect data fields and anchor elements for the entry;

generate a flexible structure description for the particular document type that includes a set of search elements for each data field in the image of the document and the title elements;

match the flexible structure description against the image; and

extract data from the image based upon the matching of the flexible structure description against the image.

10. The system of claim 9 , wherein the one or more electronic processors are further configured to adjust the flexible structure description based on user corrections of detected data fields, title elements, and/or anchor elements.

11. The system of claim 9 , wherein the table spans multiple pages of the document, and wherein the title elements repeat on two or more pages of the multiple pages.

12. The system of claim 9 , wherein to generate a flexible structure description for the particular document type the one or more electronic processors are configured to identify anchor elements for searching data fields.

13. The system of claim 9 , wherein to generate a flexible structure description for the particular document type the one or more electronic processors are configured to generate values of search parameters for each element and field.

14. The system of claim 9 , wherein to match the flexible structure description against the image comprises the one or more electronic processors are further configured to:

match the title elements on each page of the document;

remove the matched title elements from a search area of the document; and

search the search area with the matched title elements removed for data fields.

15. The system of claim 9 , wherein the entry corresponds to a row of the table that spans multiple lines of the document.

16. The system of claim 9 , wherein the one or more electronic processors are further configured to receive at least one entry, wherein the entry comprises multiple cells of the table, and wherein at least two of the multiple cells overlap.

17. A non-transitory computer-readable medium having instructions stored thereon to create a flexible structure description, the instructions comprising:

instructions to receive an image of a document of a particular document type that contains a table;

instructions to receive an entry describing an item in the table;

instructions to search for title elements based upon the entry;

instructions to detect data fields and anchor elements for the entry;

instructions to generate a flexible structure description for the particular document type that includes a set of search elements for each data field in the image of the document and the title elements;

instructions to match the flexible structure description against the image; and

instructions to extract data from the image based upon the matching of the flexible structure description against the image.

18. The non-transitory computer-readable medium of claim 17 , wherein the instructions further comprise instructions to adjust the flexible structure description based on user corrections of detected data fields, title elements, and/or anchor elements.

19. The non-transitory computer-readable medium of claim 17 , wherein the table spans multiple pages of the document, and wherein the title elements repeat on two or more pages of the multiple pages.

20. The non-transitory computer-readable medium of claim 17 , wherein the instructions to generate a flexible structure description for the particular document type comprise instructions to identify anchor elements for searching data fields.