Automated document processing system
Disclosed is a document processing system that can be used to automatically detect and process an invoice for a business entity. The system receives a document and determines the structural information from the document. The structural information includes a template object and its location on the document. The system further retrieves structural information of documents stored in the database and compares the structural information of retrieved documents to the received document based on matching criteria such as a template object name and its location. If a match is found within a predefined threshold range, the data is extracted from the received document and the structural information is extracted from the matching document and the received document is processed by applying the extracted data to the corresponding extracted structural information. The processed document is transmitted for a user review or for processing by an accounting or bookkeeping software application.
1. A computer implemented method for automating document processing, the method comprising:
receiving a first plurality of documents, each document having a format specific for a particular entity, at least two entities unrelated to each other;
determining structural information of each respective document of the first plurality by detecting a respective template object and its respective location on the respective document;
retrieving, from a database, respective structural information of each of a second plurality of documents, wherein the respective structural information includes a respective template object and its respective location on its respective document; and
for each received document of the first plurality:
comparing, based on a matching criterion, the respective location of the template object from the structural information of the received document to each of the template objects from the structural information of each retrieved document of the second plurality, to find a potential matching document;
in response to finding, based on the comparing, a document that matches the received document within a predefined threshold range, extracting data from the received document and additional structural information from the matched document to process the received document; and
transmitting the processed document to an entity for review and approval.
2. The method of claim 1 , further comprising extracting additional structural information of the received document in response to not finding a document that matches the received document within a predefined threshold range.
3. The method of claim 2 , further comprising identifying a new template object from the received document.
4. The method of claim 1 , further comprising applying the data extracted from the received document to a corresponding field extracted from the matched document.
5. The method of claim 1 , further comprising determining an amount to be paid to at least one of a business entity or a vendor on a date indicated by a due date.
6. The method of claim 1 , wherein the detecting of the respective location of a template object further includes determining co-ordinate values of the respective location.
7. The method of claim 1 , further comprising determining that the respective structural information of each retrieved document of the second plurality reaches a predefined first threshold success match when compared to the determined structural information of the received document.
8. The method of claim 7 , further comprising adding each retrieved document of the second plurality which are determined to be a successful match to a list of potential matches.
9. The method of claim 8 , further comprising determining that the structural information of a document retrieved from the list of potential matches reaches a predefined second threshold success match when compared to the structural information of the received document.
10. The method of claim 1 , wherein approval metrics are stored in a template term database, wherein the approval metrics include an approved set of template objects and related information based on the review of the processed document.
11. The method of claim 1 , further comprising identifying a graphic symbol of an entity as matching criteria for the received document.
12. The method of claim 11 wherein the graphic symbol is described by a set of descriptors and homographic information related to the bitmap of the graphic symbol.
13. A non-transitory computer readable medium comprising stored instructions executable by a processor, the instructions when executed causes the processor to:
receiving a first plurality of documents, each document having a format specific for a particular entity, at least two entities unrelated to each other;
determining structural information of each respective document of the first plurality by detecting a respective template object and its respective location on the respective document;
retrieving, from a database, respective structural information of each of a second plurality of documents, wherein the respective structural information includes a respective template object and its respective location on its respective document; and
for each received document of the first plurality:
comparing, based on a matching criterion, the respective location of the template object from the structural information of the received document to each of the template objects from the structural information of each retrieved document of the second plurality, to find a potential matching document;
in response to finding, based on the comparing, a document that matches the received document within a predefined threshold range, extracting data from the received document and additional structural information from the matched document to process the received document; and
transmitting the processed document to an entity for review and approval.
14. The computer readable medium of claim 13 , further comprising instructions that when executed cause the processor to extract additional structural information of the received document in response to not finding a matching document.
15. The computer readable medium of claim 14 , further comprising instructions that when executed cause the processor to identify a new template object from the received document.
16. The computer readable medium of claim 13 , further comprising instructions that when executed cause the processor to apply the data extracted from the received document to a corresponding field extracted from the matched document.
17. The computer readable medium of claim 13 , further comprising instructions that when executed cause the processor to determine an amount to be paid to at least one of a business entity or a vendor on a date indicated by a due date.
18. The computer readable medium of claim 13 , further comprising instructions that when executed cause the processor to determine that the respective structural information of each retrieved document of the second plurality reaches a predefined first threshold success match when compared to the determined structural information of the received document.
19. The computer readable medium of claim 18 , further comprising instructions that when executed cause the processor to add each retrieved document of the second plurality which are determined to be a successful match to a list of potential matches.
20. The computer readable medium of claim 19 , further comprising instructions that when executed cause the processor to determine that the structural information of a document retrieved from the list of potential matches reaches a predefined second threshold success match when compared to the structural information of the received document.