System and method for automated electronic catalogue management and electronic image quality assessment

Patent metadata

Field	Value
Publication number	US-11599983-B2
Application number	US-202117493417-A
Country	US
Kind code	B2
Filing date	Oct 4, 2021
Priority date	Aug 23, 2018
Publication date	Mar 7, 2023
Grant date	Mar 7, 2023

Abstract

In various examples, a system receives image data characterizing an image of an item. Additionally, the system implements a first set of operations and a second set of operations. In some examples, the first set of operations includes performing a structural similarity analysis of the item, based on the image data, and determining a structural similarity score based on the structural similarity analysis of the item. In other examples, the second set of operations includes generating a plurality of derivative images by applying a plurality of distortions to the image of the item, extracting one or more features based at least on the plurality of derivative images, and determining the quality of the image based at least on the extracted one or more features and the structural similarity score.

First claim

We claim:

1. A system comprising:
one or more processors; and
a memory resource storing a set of instructions, that when executed by the one or more processors, causes the one or more processors to:
receive image data characterizing an image of an item;
implement a first set of operations, the first set of operations including:
performing a structural similarity analysis of the item, based on the image data; and
determining a structural similarity score based on the structural similarity analysis of the item;

implement a second set of operations, the second set of operations including:
generating a plurality of derivative images by applying a plurality of distortions to the image of the item;
extracting one or more features based at least on the plurality of derivative images; and

determine a quality of the image based at least on the extracted one or more features and the structural similarity score.

2. The system of claim 1 , wherein the first set of operations and the second set of operations are implemented simultaneously.

3. The system of claim 1 , wherein extracting the one or more features is further based on the image of the image data.

4. The system of claim 1 , wherein determining the quality of the image includes:
applying a regression model to the extracted one or more features and the structural similarity score.

5. The system of claim 4 , wherein the regression model is a ridge regression model.

6. The system of claim 4 , wherein applying the regression model includes:
training a convolution neural network using the at least one of the extracted one or more features; and
utilizing the trained convolution neural network when applying the regression model to the extracted one or more features and the structural similarity score.

7. The system of claim 1 , wherein the plurality of distortions includes a mean blur, a Gaussian blur, and a bilateral blur.

8. The system of claim 1 , wherein the extracted one or more features are common across the image of the image data and the plurality of derivative images.

9. The system of claim 1 , wherein the at least one of the extracted one or more features are identified as associated with only the image of the image data.

10. The system of claim 1 , wherein performing the structural similarity analysis includes determining changes in at least luminance, contrast, and structure of the image of the image data.

11. A computer-implemented method comprising:
receiving, via a processor, image data characterizing an image of an item;
implement, via the processor, a first set of operations, the first set of operations including:
performing, via the processor, a structural similarity analysis of the item, based on the image data; and
determining, via the processor, a structural similarity score based on the structural similarity analysis of the item;

implement, via the processor, a second set of operations, the second set of operations including:
generating, via the processor, a plurality of derivative images by applying a plurality of distortions to the image of the item;
extracting, via the processor, one or more features based at least on the plurality of derivative images; and

determine, via the processor, a quality of the image based at least on the extracted one or more features and the structural similarity score.

12. The computer-implemented method of claim 11 , wherein the first set of operations and the second set of operations are implemented simultaneously.

13. The computer-implemented method of claim 11 , wherein extracting the one or more features is further based on the image of the image data.

14. The computer-implemented method of claim 11 , wherein determining the quality of the image includes:
applying a regression model to the extracted one or more features and the structural similarity score.

15. The computer-implemented method of claim 14 , wherein the regression model is a ridge regression model.

16. The computer-implemented method of claim 14 , wherein applying the regression model includes:
training a convolution neural network using the at least one of the extracted one or more features; and
utilizing the trained convolution neural network when applying the regression model to the extracted one or more features and the structural similarity score.

17. The computer-implemented method of claim 11 , wherein the plurality of distortions includes a mean blur, a Gaussian blur, and a bilateral blur.

18. The computer-implemented method of claim 11 , wherein the extracted one or more features are common across the image of the image data and the plurality of derivative images.

19. The computer-implemented method of claim 11 , wherein the at least one of the extracted one or more features are identified as associated with only the image of the image data.

20. A non-transitory computer-readable medium storing instructions, that when executed by one or more processors, causes the one or more processors to:
receive image data characterizing an image of an item;
implement a first set of operations, the first set of operations including:
performing a structural similarity analysis of the item, based on the image data; and
determining a structural similarity score based on the structural similarity analysis of the item;

implement a second set of operations, the second set of operations including:
generating a plurality of derivative images by applying a plurality of distortions to the image of the item;
extracting one or more features based at least on the plurality of derivative images; and

determine a quality of the image based at least on the extracted one or more features and the structural similarity score.

Assignees

• Walmart Apollo Llc

Inventors

• Garlapati Mani Kanteswara
• Chakraborty Souradip
• Bhat Rajesh Shreedhar

Classifications

• G06F18/28

  Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries · CPC title

• G06Q30/0603 Primary

  Catalogue creation or management · CPC title

• G06V20/00

  Scenes; Scene-specific elements (control of digital cameras H04N23/60) · CPC title

• G06V10/7715

  Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods · CPC title

• G06T2207/20081

  Training; Learning · CPC title