System for dynamic optical character recognition tuning

• US 10,192,127 B1
• Filed: 07/24/2017
• Issued: 01/29/2019
• Est. Priority Date: 07/24/2017
• Status: Active Grant

• Current Assignee: Bank of America Corp.
• Original Assignee: Bank of America Corp.
• Inventors: Hall, John B., Pepe, Jr., Michael J., Santhanam, Murali, Simpkins, Kerry Kurt
• Primary Examiner(s): Wu, Jingge
• Application Number:
  US15/658,061
• Publication Number:
  US 20190026579A1
• Time in Patent Office: 554 Days
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/127 with the intervention of an operator
• CPC Class Codes: G06V 30/1456 based on user interactions
• CPC Class Codes: G06V 30/2253 Recognition of characters printed with magnetic ink G06V30/2247 takes precedence
• CPC Class Codes: G06V 30/412 Layout analysis of documents structured with printed lines or input boxes, e.g. business forms or tables

Abstract

Embodiments of the present invention provide a system for dynamically tuning optical character recognition (OCR) processes. The system receives or captures an image of a resource document and uses a general or default OCR process to identify a source of the document and values of multiple data fields in the image of the document. When the system determines that a data field is missing or cannot be extracted, it causes a computing device to display the image of the resource document and requests user input of a coordinate area of the missing data field from an associated specialist. Once the user input is received, the system applies a data field-specific OCR process on the coordinate area of the missing data field to extract the value of the data field. This value of the missing data field can be transmitted to a processing system for further processing.

Claims

1. A system for dynamically tuning optical character recognition processes, the system comprising:

a memory device; and

a processing device operatively coupled to the memory device, wherein the processing device is configured to execute computer-readable program code to;

receive an image of a resource document comprising image coordinates associated with the resource document;

apply a general optical character recognition process to the image of the resource document to identify a resource document source or a resource document type;

identify an expected image coordinate area of the image of the resource document associated with a data field of the resource document;

apply a data field-specific optical character recognition process to the expected image coordinate area of the image of the resource document;

determine that the data field is missing from the expected image coordinate area of the image of the resource document;

in response to determining that the data field is missing, cause a user interface of a computing device to display the image of the resource document;

receive, from the user interface of the computing device, a user input associated with an updated expected image coordinate area for the data field;

apply the data field-specific optical character recognition process to the updated expected image coordinate area for the data field within the image of the resource document to extract a value of the data field;

replace the expected image coordinate area in a database with the updated expected image coordinate area; and

associate the stored updated expected image coordinate area for the data field with the identified resource document source or the resource document type.

2. The system of claim 1, wherein the user input from the user interface of the computing device comprises new image coordinate boundaries associated with the updated expected image coordinate area for the data field.

3. The system of claim 2, wherein the new image coordinate boundaries define the updated expected image coordinate area for the data field.

4. The system of claim 2, wherein the new image coordinate boundaries define an additional expected image coordinate area that is added to the expected image coordinate area for the data field to generate the updated expected image coordinate area for the data field.

5. The system of claim 2, wherein the updated expected image coordinate area comprises an area within the new image coordinate boundaries and a buffer area surrounding the new image coordinate boundaries.

6. The system of claim 1, wherein determining that the data field is missing from the expected image coordinate area of the image of the resource document further comprises determining that the data field (i) is not found in the expected image coordinate area or (ii) could not be properly extracted by the data field-specific optical character recognition process.

7. The system of claim 1, wherein the processing device is further configured to execute the computer-readable program code to:

transmit the value of the data field to a resource processing system in response to extracting the value of the data field.

8. The system of claim 1, wherein the processing device is further configure to execute the computer-readable program code to:

receive an image of a new resource document comprising image coordinates associated with the new resource document;

apply the general optical character recognition process to the image of the resource document;

determine, based on the general optical character recognition process, that the new resource document is associated with the identified resource document source or resource document type; and

apply the data field-specific optical character recognition process to the updated expected image coordinate area for the data field within the image of the new resource document to extract a value of the data field for the new resource document.

9. A computer program product for dynamically tuning optical character recognition processes, the computer program product comprising at least one non-transitory computer readable medium comprising computer readable instructions that, when executed, perform steps of:

receiving an image of a resource document comprising image coordinates associated with the resource document;

applying a general optical character recognition process to the image of the resource document to identify a resource document source or a resource document type;

identifying an expected image coordinate area of the image of the resource document associated with a data field of the resource document;

applying a data field-specific optical character recognition process to the expected image coordinate area of the image of the resource document;

determining that the data field is missing from the expected image coordinate area of the image of the resource document;

in response to determining that the data field is missing, causing a user interface of a computing device to display the image of the resource document;

receiving, from the user interface of the computing device, a user input associated with an updated expected image coordinate area for the data field;

applying the data field-specific optical character recognition process to the updated expected image coordinate area for the data field within the image of the resource document to extract a value of the data field;

replacing the expected image coordinate area in a database with the updated expected image coordinate area; and

associating the stored updated expected image coordinate area for the data field with the identified resource document source or the resource document type.

10. The computer program product of claim 9, wherein the user input from the user interface of the computing device comprises new image coordinate boundaries associated with the updated expected image coordinate area for the data field.

11. The computer program product of claim 10, wherein the new image coordinate boundaries define the updated expected image coordinate area for the data field.

12. The computer program product of claim 10, wherein the new image coordinate boundaries define an additional expected image coordinate area that is added to the expected image coordinate area for the data field to generate the updated expected image coordinate area for the data field.

13. The computer program product of claim 10, wherein the updated expected image coordinate area comprises an area within the new image coordinate boundaries and a buffer area surrounding the new image coordinate boundaries.

14. The computer program product of claim 9, wherein determining that the data field is missing from the expected image coordinate area of the image of the resource document further comprises determining that the data field (i) is not found in the expected image coordinate area or (ii) could not be properly extracted by the data field-specific optical character recognition process.

15. The computer program product of claim 9, wherein the computer readable instructions further comprise instructions that, when executed, performs a step of:

transmitting the value of the data field to a resource processing system in response to extracting the value of the data field.

16. The computer program product of claim 9, wherein the computer readable instructions further comprise instructions that, when executed, perform steps of:

receiving an image of a new resource document comprising image coordinates associated with the new resource document;

applying the general optical character recognition process to the image of the resource document;

determining, based on the general optical character recognition process, that the new resource document is associated with the identified resource document source or resource document type; and

applying the data field-specific optical character recognition process to the updated expected image coordinate area for the data field within the image of the new resource document to extract a value of the data field for the new resource document.

17. A computer implemented method for dynamically tuning optical character recognition processes, said computer implemented method comprising:

providing a computing system comprising a computer processing device and a non-transitory computer readable medium, where the computer readable medium comprises configured computer program instruction code, such that when said instruction code is operated by said computer processing device, said computer processing device performs the following operations;

receiving an image of a resource document comprising image coordinates associated with the resource document;

applying a general optical character recognition process to the image of the resource document to identify a resource document source or a resource document type;

identifying an expected image coordinate area of the image of the resource document associated with a data field of the resource document;

applying a data field-specific optical character recognition process to the expected image coordinate area of the image of the resource document;

determining that the data field is missing from the expected image coordinate area of the image of the resource document;

in response to determining that the data field is missing, causing a user interface of a computing device to display the image of the resource document;

receiving, from the user interface of the computing device, a user input associated with an updated expected image coordinate area for the data field;

applying the data field-specific optical character recognition process to the updated expected image coordinate area for the data field within the image of the resource document to extract a value of the data field;

replacing the expected image coordinate area in a database with the updated expected image coordinate area; and

associating the stored updated expected image coordinate area for the data field with the identified resource document source or the resource document type.

18. The computer implemented method of claim 17, wherein the user input from the user interface of the computing device comprises new image coordinate boundaries associated with the updated expected image coordinate area for the data field, and wherein the new image coordinate boundaries define the updated expected image coordinate area for the data field.

19. The computer implemented method of claim 17, wherein the user input from the user interface of the computing device comprises new image coordinate boundaries associated with the updated expected image coordinate area for the data field, and wherein the new image coordinate boundaries define an additional expected image coordinate area that is added to the expected image coordinate area for the data field to generate the updated expected image coordinate area for the data field.

20. The computer implemented method of claim 17, wherein the user input from the user interface of the computing device comprises new image coordinate boundaries associated with the updated expected image coordinate area for the data field, and wherein the updated expected image coordinate area comprises an area within the new image coordinate boundaries and a buffer area surrounding the new image coordinate boundaries.