Granted Patent B2

US 12,430,879 · App. 17/949,585 · Granted Sep 30, 2025

Inventor: Renjith Radhakrishna Pillai Suseelamma (Kollam, IN)

Assignee: HRB Innovations, Inc.

G06V10/759G06V10/449G06V10/762G06V20/60H04N1/40012

Abstract

Claims (62)

1. One or more non-transitory computer-readable media storing computer-executable instructions that, when executed by one or more processors, perform a method of comparing images, the method comprising:

receiving a first image to be compared and a second image to be compared;

converting the first image to be compared and the second image to be compared to grayscale;

identifying a first plurality of unmatched regions in the first image to be compared;

identifying a second plurality of unmatched regions in the second image to be compared;

identifying each first unmatched region in the first plurality of unmatched regions with a corresponding second unmatched region in the second plurality of unmatched regions to obtain a plurality of image differences;

for each image difference in the plurality of image differences, calculating a corresponding logical match percentage;

for each image difference in the plurality of image differences, determining whether the image difference is a significant image difference based on the corresponding logical match percentage to obtain one or more significant image differences;

generating a comparison result image illustrating the one or more significant image differences; and

generating an image match report including the comparison result image illustrating the one or more significant image differences.

2. The non-transitory computer-readable media of claim 1 , the method further comprising:

comparing image properties between the first image to be compared and the second image to be compared; and

in response to identifying different image properties between the first image to be compared and the second image to be compared, adjusting the image properties of at least one of the first image to be compared and the second image to be compared.

3. The non-transitory computer-readable media of claim 1 , wherein identifying each first unmatched region in the first plurality of unmatched regions with the corresponding second unmatched region in the second plurality of unmatched regions comprises:

performing cluster analysis of the first plurality of unmatched regions and the second plurality of unmatched regions;

assigning a first unmatched region of the first plurality of unmatched regions into a first cluster and a second unmatched region of the second plurality of unmatched regions into a second cluster; and

comparing the first unmatched region to a plurality of candidate unmatched regions from the second plurality of unmatched regions based on cluster analysis results from the first cluster and the second cluster.

4. The non-transitory computer-readable media of claim 1 , wherein identifying each first unmatched region in the first plurality of unmatched regions with the corresponding second unmatched region in the second plurality of unmatched regions comprises:

identifying an instance of data sliding between the first unmatched region and the corresponding second unmatched region.

5. The non-transitory computer-readable media of claim 1 , the method further comprising:

identifying a novel object in the second image to be compared relative to the first image to be compared.

6. The non-transitory computer-readable media of claim 1 , wherein identifying each first unmatched region in the first plurality of unmatched regions with the corresponding second unmatched region in the second plurality of unmatched regions comprises:

performing a first blob detection on the first unmatched region;

performing a second blob detection on the corresponding second unmatched region; and

comparing one or more first blobs identified in the first blob detection with one or more blobs identified in the second blob detection.

7. The non-transitory computer-readable media of claim 6 , wherein performing blob detection on the first unmatched region comprises:

using a blob detection technique selected from a set consisting of: a Laplacian-of-Gaussian algorithm, a Difference-of-Gaussian algorithm, and a Determinant-of-Hessian algorithm.

8. A method of comparing images, comprising:

receiving a first image to be compared and a second image to be compared;

identifying a first plurality of unmatched regions in the first image to be compared;

identifying a second plurality of unmatched regions in the second image to be compared;

performing cluster analysis of the first plurality of unmatched regions and the second plurality of unmatched regions;

identifying each first unmatched region in the first plurality of unmatched regions with a corresponding second unmatched region in the second plurality of unmatched regions to obtain a plurality of image differences;

for each image difference in the plurality of image differences, calculating a corresponding logical match percentage,

for each image difference in the plurality of image differences, determining whether the image difference is a significant image difference based on the corresponding logical match percentage to obtain one or more significant image differences;

generating a comparison result image illustrating the one or more significant image differences; and

generating an image match report including the comparison result image illustrating the one or more significant image differences.

9. The method of claim 8 , wherein the corresponding logical match percentage is based at least in part on a Euclidean Distance Logic calculation for the first unmatched region and the corresponding second unmatched region.

10. The method of claim 9 , wherein the corresponding logical match percentage is based in further part on a mode pixel value comparison between the first unmatched region and the corresponding second unmatched region.

11. The method of claim 8 ,

wherein the image match report further includes a plurality of corresponding logical match percentages.

12. The method of claim 8 , wherein each region in the first plurality of unmatched regions and the second plurality of unmatched regions corresponds to a region selected from a set consisting of: a text region, an icon region, and a shape region.

13. The method of claim 8 , wherein each region in the first plurality of unmatched regions and the second plurality of unmatched regions corresponds to a layer selected from a set consisting of: a foreground layer and a background layer.

14. The method of claim 8 , further comprising:

converting the first image to be compared and the second image to be compared to grayscale.

15. A system for comparing images comprising:

one or more processors; and

at least one non-transitory memory storing computer executable instructions that, when executed by the one or more processors, perform a method of comparing the images, the method comprising:

receiving a first image to be compared and a second image to be compared;

identifying a first plurality of unmatched regions in the first image to be compared;

identifying a second plurality of unmatched regions in the second image to be compared;

identifying each first unmatched region in the first plurality of unmatched regions with a corresponding second unmatched region in the second plurality of unmatched regions to obtain a plurality of image differences;

for each image difference in the plurality of image differences, calculating a corresponding logical match percentage;

for each image difference in the plurality of image differences, determining whether the image difference is significant to identify one or more significant image differences;

generating a comparison result image illustrating the one or more significant image differences; and

generating an image match report including the comparison result image illustrating the one or more significant image differences.

16. The system of claim 15 , wherein the method further comprises:

identifying an instance of data sliding between the first unmatched region and the corresponding second unmatched region.

17. The system of claim 15 , wherein the image match report includes a plurality of logical match percentages and a plurality of physical match percentages.

18. The system of claim 15 , wherein identifying each first unmatched region in the first plurality of unmatched regions with the corresponding second unmatched region in the second plurality of unmatched regions comprises a using technique selected from a set consisting of: a Laplacian-of-Gaussian algorithm, a Difference-of-Gaussian algorithm, and a Determinant-of-Hessian algorithm.

19. The system of claim 15 , wherein the image match report excludes at least one image difference determined not to be significant.

20. The system of claim 15 , wherein the image match report includes screenshots illustrating particular instances of significant image differences.

Assignments (1)

Continuity (1)

References Cited (15)