Compensation for print shift in standardized forms to facilitate extraction of data therefrom

• US 9,411,547 B1
• Filed: 07/28/2015
• Issued: 08/09/2016
• Est. Priority Date: 07/28/2015
• Status: Active Grant

• Current Assignee: DST Technologies, Inc. (SS&C Technologies Holdings Incorporated)
• Original Assignee: DST Technologies, Inc. (SS&C Technologies Holdings Incorporated)
• Inventors: Bessinger, Keith, Kirchmeier, John
• Primary Examiner(s): Nguyen, Ngon
• Application Number:
  US14/811,069
• Time in Patent Office: 378 Days
• Field of Search: 358/1.12
• US Class Current: 1/1
• CPC Class Codes: G06F 40/174 Form filling; Merging

Abstract

A method comprising receiving a candidate image, including a first field, a second field, a first data element, and a second data element. The processor receives a definition of first and second form zones. The processor evaluates the first data element to determine a first measure of a similarity and the second data element to determine a second measure of a similarity. The processor computes a similarity value indicative of an overall similarity, iteratively compares the overall similarity to a previous iteration, modifies the positions of the form zones, and repeats the evaluating and computing based on the modified positions until the computed similarity value meets a threshold.

Claims

1. A computer implemented method comprising:

receiving, by a processor, a candidate image, wherein the candidate image contains a first field and a first data element positioned in a first position with respect thereto within the candidate image, and a second field and a second data element positioned in a second position with respect thereto within the candidate image, and storing the candidate image in a memory coupled with the processor;

receiving, by the processor, a definition of first and second form zones, and storing the definition of the first and second form zones in a memory coupled with the processor, wherein the first form zone defines a third position at which the first data element is expected to be positioned with respect to the first field and the second form zone defines a fourth position at which the second data element is expected to be positioned with respect to the second field;

evaluating, by the processor, the first data element with respect to the third position to determine a first measure of a similarity of the first position to the third position and the second data element with respect to the fourth position to determine a second measure of a similarity of the second position to the fourth position;

computing, by the processor, a similarity value indicative of an overall similarity of the first and second positions to the third and fourth positions based on the first and second measures; and

iteratively comparing, by the processor, the similarity value to a similarity value of a previous iteration, modifying, by the processor, the third and fourth positions and repeating the evaluating and computing based on the modified third and fourth positions until the computed similarity value meets a threshold.

2. The computer implemented method of claim 1, further comprising:

extracting, by the processor, first data from the modified third position and second data from the modified fourth position; and

associating, by the processor, the extracted first data with the first field and the extracted second data with the second field.

3. The computer implemented method of claim 1, wherein modifying comprises:

incrementing, by the processor, the third and fourth positions by a shift vector stored in the memory.

4. The computer implemented method of claim 3, wherein a shift vector is generated by a processor by incrementing a previously selected shift vector by a number of pixels.

5. The computer implemented method of claim 3, further comprising:

identifying, by the processor, a form type of the candidate image; and

selecting, by the processor, a shift vector that corresponds to the form type.

6. The computer implemented method of claim 1, wherein evaluating the first data element with respect to the third position to determine a first measure of a similarity comprises:

calculating, by the processor, an overlap between the first position of the data element and the third position.

7. The computer implemented method of claim 1, wherein receiving a candidate image further comprises:

identifying, by the processor, the first position of the first data element and the second position of the second data element.

8. The computer implemented method of claim 7, wherein the candidate image contains form data;

and wherein identifying the first position and the second position uses drop out ink to separate the form data from the first and second data elements.

9. The computer implemented method of claim 7, wherein the candidate image contains form data;

and wherein identifying the first position and the second position uses form template identification to separate the form data from the first and second data elements.

10. The computer implemented method of claim 1, wherein computing the similarity value based on the first and second measures comprises averaging the first and second measures.

11. The computer implemented method of claim 1, wherein computing the similarity value based on the first and second measures comprises summing the first and second measures.

12. The computer implemented method of claim 1, wherein the computed similarity value meets the threshold when the third position coincides with the first position and the fourth position coincides with the second position.

13. A system comprising:

a controller in communication with a storage device configured to receive and accessibly store a generated one or more candidate images and candidate image data, wherein the controller is operable to;

receive a candidate image, wherein the candidate image contains a first field, a second field, a first data element, and a second data element;

identify a first position for the first data element with respect thereto within the candidate image, and a second position for the second data element with respect thereto within the candidate image;

receive a definition of first and second form zones, wherein the first form zone defines a third location at which the first data element is expected to be positioned with respect to the first field and the second form zone defines a fourth location at which the second data element is expected to be positioned with respect to the second field;

evaluate the first data element with respect to the third position to determine a first measure of a similarity of the first position to the third position and the second data element with respect to the fourth position to determine a second measure of a similarity of the second position to the fourth position;

compute a similarity value indicative of an overall similarity of the first and second positions to the third and fourth positions based on the first and second measures; and

iteratively compare the overall similarity to a previous iteration, modify the third and fourth positions and repeat the evaluating and computing based on the modified third and fourth positions until the computed similarity value meets a threshold.

14. The system of claim 13, wherein the controller is further operable to:

extract data from the third position of the modified first form zone; and

associate the extracted data with the first field.

15. The system of claim 13, wherein a measure of a similarity is a percentage of the first position of the first data element that overlap with the third position of the first form zone.

16. The system of claim 13, wherein the controller is further operable to:

modify the third and fourth positions by a shift vector, wherein the shift vector comprise a horizontal number of pixels and a vertical number of pixels.

17. The system of claim 13, wherein the candidate image includes form data;

and

wherein the controller is further operable to;

identify the first position and the second position by using drop out ink to separate the form data from the first and second data elements.

18. The system of claim 13, wherein the computed similarity value meets the threshold when the third position coincides with the first position and the fourth position coincides with the second position.

19. The system of claim 13, wherein the controller is further operable to compute the similarity value by averaging the first and second measures.

20. A system comprising:

means for receiving a candidate image, wherein the candidate image contains a first field and a first data element positioned in a first position with respect thereto within the candidate image, and a second field and a second data element positioned in a second position with respect thereto within the candidate image;

means for receiving a definition of first and second form zones, wherein the first form zone defines a third position at which the first data element is expected to be positioned with respect to the first field and the second form zone defines a fourth position at which the second data element is expected to be positioned with respect to the second field;

means for evaluating the first data element with respect to the third position to determine a first measure of a similarity of the first position to the third position and the second data element with respect to the fourth position to determine a second measure of a similarity of the second position to the fourth position;

means for computing a similarity value indicative of an overall similarity of the first and second positions to the third and fourth positions based on the first and second measures; and

means for iteratively comparing the similarity value to a similarity value of a previous iteration, modifying the third and fourth positions and repeating the evaluating and computing based on the modified third and fourth positions until the computed similarity value meets a threshold.