OPTICAL CHARACTER RECOGNITIONS VIA CONSENSUS OF DATASETS

• US 20200134348A1
• Filed: 07/21/2017
• Published: 04/30/2020
• Est. Priority Date: 07/21/2017
• Status: Active Grant

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Breslav, Mikhail
• Granted Patent: US 11,328,167 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 18/254 of classification results, e.g. of results related to same input data
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/12 Detection or correction of errors, e.g. by rescanning the pattern
• CPC Class Codes: G06V 30/1918 Fusion techniques, i.e. combining data from various sources, e.g. sensor fusion
• CPC Class Codes: G06V 30/418 Document matching, e.g. of document images

Abstract

An example of apparatus includes a memory to store a first image of a document and a second image of the document. The first image and the second image are Memory captured under different conditions. The apparatus includes a processor coupled to the memory. The processor is to perform optical character recognition on the first image to generate a first output dataset and to perform optical character recognition on the second image to generate a second output dataset. The processor is further to determine whether consensus for a character is achieved based on a comparison of the first output dataset with the second output dataset, and generate a final output dataset based on the consensus for the character.

Claims

1. An apparatus comprising:

a memory to store a first image of a document and a second image of the document, wherein the first image and the second image are captured under different conditions; and

a processor coupled to the memory, wherein the processor is to;

perform optical character recognition on the first image to generate a first output dataset;

perform optical character recognition on the second image to generate a second output dataset;

determine whether consensus for a character is achieved based on a comparison of the first output dataset with the second output dataset; and

generate a final output dataset based on the consensus for the character.

2. The apparatus of claim 1, wherein the processor is to perform an alignment based on the first output dataset and the second output dataset.

3. The apparatus of claim 1, wherein the processor is to define a voting position, wherein the processor is to select a first region from the first image to be associated with the voting position, and wherein the processor is to select a second region from the second image to be associated with the voting position.

4. The apparatus of claim 3, wherein the first region is based on the first output dataset.

5. The apparatus of claim 4, wherein the second region is based on the second output dataset.

6. The apparatus of claim 5, wherein the processor is to count occurrences of a character associated with the voting position to determine whether the consensus is achieved.

7. The apparatus of claim 6, wherein the processor is to apply a weighting scheme to the first output dataset and the second output dataset.

8. A non-transitory machine-readable storage medium encoded with instructions executable by a processor, the non-transitory machine-readable storage medium comprising:

instructions to generate a first plurality of characters from optical character recognition of a first image;

instructions to generate a second plurality of characters from optical character recognition of a second image;

instructions to count occurrences of a character in the first plurality of characters and the second plurality of characters to establish a majority; and

instructions to generate a final output dataset based on the majority for the character.

9. The non-transitory machine-readable storage medium of claim 8, comprising instructions to capture the first image using an image sensor.

10. The non-transitory machine-readable storage medium of claim 9, comprising instructions to capture the second image using an image sensor.

11. The non-transitory machine-readable storage medium of claim 8, comprising instructions to define a voting position, wherein the processor selects a first region from the first image to be associated with the voting position and the processor selects a second region from the second image to be associated with the voting position.

12. The non-transitory machine-readable storage medium of claim 11, comprising instructions to count occurrences of a character associated with the voting position to determine whether the majority is achieved.

13. An apparatus comprising:

a communications interface to communicate with a network;

a user interface to receive input from a user and to generate output;

an image sensor;

a memory to store a first image and a second image, wherein the first image and the second image are images of a document captured under different conditions, wherein the first image and the second image are received via one of the communication interface or the image sensor; and

a processor coupled to the memory, the communication interface, the user interface, and the image sensor, wherein the processor is to;

perform optical character recognition on the first image to generate a first plurality of characters;

perform optical character recognition on the second image to generate a second plurality of characters;

determine whether consensus for a character is achieved based on a comparison of the first plurality of characters with the second plurality of characters; and

generate a final output dataset based on the consensus for the character.

14. The apparatus of claim 13, wherein the processor is to perform an alignment based on the first plurality of characters and the second plurality of characters.

15. The apparatus of claim 14, wherein the processor is to apply a weighting scheme to the first plurality of characters and the second plurality of characters.