Fraudulent activity detection at a barcode scanner by verifying visual signatures

Patent metadata

Field	Value
Publication number	US-10825027-B2
Application number	US-201916293857-A
Country	US
Kind code	B2
Filing date	Mar 6, 2019
Priority date	Jan 7, 2014
Publication date	Nov 3, 2020
Grant date	Nov 3, 2020

Abstract

System and method for detecting a fraudulent activity at a barcode scanner is disclosed. The method issues an alert when the fraudulent activity is confirmed by comparing the visual signature of the item being transacted over the checkout terminal to the model visual signature. The model visual signature is obtained by averaging the collection of visual signature of the item gathered over a period of time. A human validation via a remote processor is employed to confirm the fraudulent activity verified by a computer.

First claim

What is claimed is:

1. A method, comprising:
capturing, by a camera, an image representing a portion of an item, wherein capturing further includes capturing within the portion an identifier for the item along with other areas that surround the identifier and that are part of packaging for the item but are not part of the identifier, the other areas excluding the identifier, wherein the portion comprises the identifier and the other areas;
obtaining, by executable instructions that execute on a processor from a non-transitory computer-readable storage medium, the identifier for the item;
acquiring, by the executable instructions, a signature corresponding to the portion of the item that is associated with the other areas;
calculating, by the executable instructions, an image signature for the image;
comparing, by the executable instructions, the signature against the image signature; and
issuing, by the executable instructions, an alert to an alert system based on the comparing.

2. The method of claim 1 , wherein obtaining further includes determining the item identifier based on the image.

3. The method of claim 2 , wherein determining further includes resolving the item identifier from a barcode represented within the image.

4. The method of claim 3 , wherein acquiring further includes obtaining the signature as a model signature for an area on the item associated with known packaging features of that item.

5. The method of claim 1 , wherein acquiring further includes obtaining the signature from a plurality of signatures associated with known packaging features of that item.

6. The method of claim 5 , wherein obtaining further includes selecting the signature from the plurality of signatures based on an area associated with the portion of the item represented in the image.

7. The method of claim 6 , wherein calculating further includes calculating the image signature based on selective points within the image that are associated with the other areas.

8. The method of claim 1 , wherein calculating includes calculating the image signature based on selective points identified within the image that are associated with the other areas.

9. The method of claim 8 , wherein calculating further includes selecting the selective points as points within the portion of the image that surround and are outside of a barcode represented within the image.

10. A method, comprising:
determining, by executable instructions that execute on a processor from a non-transitory computer-readable storage medium, an item identifier for an item during a transaction from a first portion of an item image captured for the item;
calculating, by the executable instructions, a signature for a second portion of the item image, wherein the second portion are other areas of the item that do not include the first portion, wherein the second portion comprising packaging features for the item, wherein the second portion excludes the first portion; and
raising, by the executable instructions, an alert when the signature fails to match a model signature associated with the second portion of the item image.

11. The method of claim 10 , wherein calculating further includes identifying the second portion as an area within the item image that surrounds the first portion.

12. The method of claim 10 , wherein calculating further includes selecting the second portion of the item image based on model signature and an area on the item from which the model signature is based.

13. The method of claim 12 , wherein selecting further includes selecting the second portion as positions within the image that do not include an item barcode, wherein the item barcode present within the first portion of the item image.

14. The method of claim 10 , wherein alerting further includes halting transaction processing associated with the transaction.

15. The method of claim 14 , wherein halting further includes providing a video representing a time frame when the item image was captured for the transaction.

16. The method of claim 15 , wherein providing further includes providing other item images captured at different angles for the item within the time frame.

17. The method of claim 15 further comprising, sending, by the executable instructions, the video to a remote agent for review of the video while the transaction remains halted.

18. The method of claim 10 , wherein calculating further includes correcting image distortions present within the second portion of the item image before calculating the signature.

19. A device, comprising:
a processor;
a non-transitory computer-readable storage medium having executable instructions; and
the executable instructions when executed by the processor from the non-transitory computer-readable storage medium cause the processor to:
identify an item from a first portion of an item image captured of the item;
obtain model signatures for the item;
select a second portion of the item image based on areas associated with the item from which the model signatures are based, wherein the second portion excludes the first portion, the second portion surrounding the first portion, and the second portion is associated with other areas of the item that do not include the first portion, wherein the second portion comprising packaging features for the item;
calculate an item signature based on at least one point represented for the item in the second portion of the item image;
identify a particular model signature based on the at least one point;
compare the particular model signature against the item signature that was calculated; and
raise at least one alert when the particular model signature does not match the item signature.

20. The device of claim 19 , wherein the first portion is an item barcode area representing an item barcode on the item.

Assignees

• Migdal Joshua
• Kundu Malay
• Ncr Corp

Inventors

• Migdal Joshua
• Kundu Malay

Classifications

• G06Q30/018 Primary

  Certifying business or products · CPC title