Computer architecture for emulating a foveal mechanism in a correlithm object processing system

• US 10,481,930 B1
• Filed: 06/25/2018
• Issued: 11/19/2019
• Est. Priority Date: 06/25/2018
• Status: Active Grant

• Current Assignee: Bank of America Corp.
• Original Assignee: Bank of America Corp.
• Inventors: Lawrence, Patrick N.
• Primary Examiner(s): Nguyen, Vu
• Application Number:
  US16/017,729
• Time in Patent Office: 512 Days
• Field of Search: None
• CPC Class Codes: G06F 16/58 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 9/45504 Abstract machines for programme code execution, e.g. Java virtual machine [JVM], interpreters, emulators
• CPC Class Codes: G06Q 10/00 Administration; Management

Abstract

a device configured to emulate a foveal mechanism in a correlithm object processing system comprises a memory and one or more processors. The memory stores a correlithm object mapping table that includes multiple source image elements and multiple corresponding target correlithm objects. The processors receive an image having a bitmap format and identify a first focus block of the image that includes multiple sub-blocks. The processors then compute a composite score for the first focus block that indicates a matching level between the first focus block and the source image. The processors then identify a direction to move the first focus block to a second focus block. The processors then compute a second composite score for the second focus block with respect to the source image and compare the second composite score to the first composite score to determine whether the matching level improves.

Claims

1. A device to emulate a foveal mechanism in a correlithm object processing system, comprising:

a memory operable to store a correlithm object mapping table that comprises a plurality of source image elements and a plurality of corresponding target correlithm objects, each of the plurality of source image elements comprising an n-pixel array of binary values; and

one or more processors communicatively coupled to the memory and configured to;

receive an image;

identify a first focus block of the image, the first focus block comprising an n-pixel array of binary values, the first focus block further comprising a plurality of sub-blocks, the plurality of sub-blocks comprising at least a central sub-block surrounded by a plurality of neighboring sub-blocks;

determine, for each of the sub-blocks in the first focus block, a matching score of the sub-block with respect to a source image element;

compute a first composite score for the first focus block by summing the matching score for each of the sub-blocks of the first focus block;

identify a neighboring sub-block of the first focus block with the highest matching score;

identify a first direction that directs from the central sub-block of the first block to the determined neighboring sub-block with the highest matching score;

identify a second focus block of the image, the second focus block of the image comprising a plurality of sub-blocks and being a predetermined distance away from the first focus block along the identified first direction;

determine, for each of the sub-blocks in the second focus block, a matching score of the sub-block with respect to the source image element;

compute a second composite score for the second focus block by summing the matching score for each of the sub-blocks of the second focus block;

compare the second composite score to the first composite score; and

in response to determining that the second composite score is less than the first composite score, determine an n-dimensional distance between the first focus block of the image and the source image element.

2. The device of claim 1, wherein the first focus block is in a central area of the image.

3. The device of claim 1, wherein the image has a format of a bitmap.

4. The device of claim 3, wherein determining, for each of the sub-blocks in the first focus block, a matching score of the sub-block with respect to a source image element comprises:

identifying a local patch of the source image element that closely matches the sub-block in the first focus block based at least in part on a feature detection method, the identified local patch of the source image element comprising an m-pixel array of binary values;

determining an m-dimensional distance between the sub-block of the first focus block and the identified local patch of the source image element; and

computing a matching score of the sub-block with respect to the source image element based on the determined m-dimensional distance, the matching score being inversely proportional to the determined m-dimensional distance.

5. The device of claim 1, wherein each sub-block of the first focus block comprises an m-pixel array of binary values, wherein the value of m is less than the value of n.

6. The device of claim 1, wherein the predetermined distance comprises one sub-block of the first focus block.

7. The device of claim 1, wherein the one or more processors are further configured to:

in response to determining that the second composite score is equal to or greater than the first composite score, identify a neighboring sub-block of the second focus block with the highest matching score;

identify a second direction that directs from a central sub-block of the second focus block to the determined neighboring sub-block with the highest matching score;

identify a third focus block of the image, the third focus block being a predetermined distance away from the second focus block along the identified second direction;

determine, for each of the sub-blocks in the third focus block, a matching score of the sub-block with respect to the source image element; and

compute a third composite score for the third focus block by summing the matching score for each of the sub-blocks of the third focus block.

8. A non-transitory computer-readable medium comprising logic for emulating a foveal mechanism in a correlithm object processing system, the logic, when executed by a processor, operable to:

store a correlithm object mapping table that comprises a plurality of source image elements and a plurality of corresponding target correlithm objects, each of the plurality of source image elements comprising an n-pixel array of binary values;

receive an image;

identify a first focus block of the image, the first focus block comprising an n-pixel array of binary values, the first focus block further comprising a plurality of sub-blocks, the plurality of sub-blocks comprising at least a central sub-block surrounded by a plurality of neighboring sub-blocks;

determine, for each of the sub-blocks in the first focus block, a matching score of the sub-block with respect to a source image element;

compute a first composite score for the first focus block by summing the matching score for each of the sub-blocks of the first focus block;

identify a neighboring sub-block of the first focus block with the highest matching score;

identify a first direction that directs from the central sub-block of the first block to the determined neighboring sub-block with the highest matching score;

identify a second focus block of the image, the second focus block of the image comprising a plurality of sub-blocks and being a predetermined distance away from the first focus block along the identified first direction;

determine, for each of the sub-blocks in the second focus block, a matching score of the sub-block with respect to the source image element;

compute a second composite score for the second focus block by summing the matching score for each of the sub-blocks of the second focus block;

compare the second composite score to the first composite score; and

in response to determining that the second composite score is less than the first composite score, determine an n-dimensional distance between the first focus block of the image and the source image element.

9. The non-transitory computer-readable medium of claim 8, wherein the first focus block is in a central area of the image.

10. The non-transitory computer-readable medium of claim 8, wherein the image has a format of a bitmap.

11. The non-transitory computer-readable medium of claim 10, wherein determining, for each of the sub-blocks in the first focus block, a matching score of the sub-block with respect to a source image element comprises:

identifying a local patch of the source image element that closely matches the sub-block in the first focus block based at least in part on a feature detection method, the identified local patch of the source image element comprising an m-pixel array of binary values;

determining an m-dimensional distance between the sub-block of the first focus block and the identified local patch of the source image element; and

computing a matching score of the sub-block with respect to the source image element based on the determined m-dimensional distance, the matching score being inversely proportional to the determined m-dimensional distance.

12. The non-transitory computer-readable medium of claim 8, wherein the predetermined distance comprises one sub-block of the first focus block.

13. The non-transitory computer-readable medium of claim 8, wherein the one or more processors are further configured to:

in response to determining that the second composite score is equal to or greater than the first composite score, identify a neighboring sub-block of the second focus block with the highest matching score;

identify a second direction that directs from a central sub-block of the second focus block to the determined neighboring sub-block with the highest matching score;

identify a third focus block of the image, the third focus block being a predetermined distance away from the second focus block along the identified second direction;

determine, for each of the sub-blocks in the third focus block, a matching score of the sub-block with respect to the source image element; and

compute a third composite score for the third focus block by summing the matching score for each of the sub-blocks of the third focus block.

14. A method for emulating a foveal mechanism in a correlithm object processing system, comprising:

storing a correlithm object mapping table that comprises a plurality of source image elements and a plurality of corresponding target correlithm objects, each of the plurality of source image elements comprising an n-pixel array of binary values;

receiving an image;

identifying a first focus block of the image, the first focus block comprising an n-pixel array of binary values, the first focus block further comprising a plurality of sub-blocks, the plurality of sub-blocks comprising at least a central sub-block surrounded by a plurality of neighboring sub-blocks;

determining, for each of the sub-blocks in the first focus block, a matching score of the sub-block with respect to a source image element;

computing a first composite score for the first focus block by summing the matching score for each of the sub-blocks of the first focus block;

identifying a neighboring sub-block of the first focus block with the highest matching score;

identifying a first direction that directs from the central sub-block of the first block to the determined neighboring sub-block with the highest matching score;

identifying a second focus block of the image, the second focus block of the image comprising a plurality of sub-blocks and being a predetermined distance away from the first focus block along the identified first direction;

determining, for each of the sub-blocks in the second focus block, a matching score of the sub-block with respect to the source image element;

computing a second composite score for the second focus block by summing the matching score for each of the sub-blocks of the second focus block;

comparing the second composite score to the first composite score; and

in response to determining that the second composite score is less than the first composite score, determining an n-dimensional distance between the first focus block of the image and the source image element.

15. The method of claim 14, wherein the first focus block is in a central area of the image.

16. The method of claim 14, wherein the image has a format of a bitmap.

17. The method of claim 16, wherein determining, for each of the sub-blocks in the first focus block, a matching score of the sub-block with respect to a source image element comprises:

identifying a local patch of the source image element that closely matches the sub-block in the first focus block based at least in part on a feature detection method, the identified local patch of the source image element comprising an m-pixel array of binary values;

determining an m-dimensional distance between the sub-block of the first focus block and the identified local patch of the source image element; and

computing a matching score of the sub-block with respect to the source image element based on the determined m-dimensional distance, the matching score being inversely proportional to the determined m-dimensional distance.

18. The device of claim 14, wherein each sub-block of the first focus block comprises an m-pixel array of binary values, wherein the value of m is less than the value of n.

19. The method of claim 14, wherein the predetermined distance comprises one sub-block of the first focus block.

20. The method of claim 14, wherein the one or more processors are further configured to:

in response to determining that the second composite score is equal to or greater than the first composite score, identify a neighboring sub-block of the second focus block with the highest matching score;

identify a second direction that directs from a central sub-block of the second focus block to the determined neighboring sub-block with the highest matching score;

identify a third focus block of the image, the third focus block being a predetermined distance away from the second focus block along the identified second direction;

determine, for each of the sub-blocks in the third focus block, a matching score of the sub-block with respect to the source image element; and

compute a third composite score for the third focus block by summing the matching score for each of the sub-blocks of the third focus block.