Using augmented reality data as part of a fraud detection process

Patent metadata

Field	Value
Publication number	US-11854014-B2
Application number	US-202016918730-A
Country	US
Kind code	B2
Filing date	Jul 1, 2020
Priority date	Jul 1, 2020
Publication date	Dec 26, 2023
Grant date	Dec 26, 2023

Abstract

A fraud detection system may identify a suspected fraudulent transaction based on one or more criteria. The transaction may be associated with a user. Based on identifying a suspected fraudulent transaction, the fraud detection system may receive augmented reality data from an augmented reality device of the user. The fraud detection system may determine, based on the augmented reality data, whether the transaction is fraudulent, and may take an action based on the determination.

First claim

What is claimed is:

1. A method comprising:
identifying, by a fraud detection server system, a transaction associated with a user, wherein the transaction is flagged as potentially fraudulent and corresponds to a transaction record comprising merchant category information;
receiving, video from a video feed of an augmented reality (AR) device associated with the user;
determining, based on the transaction record comprising the merchant category information, an environment associated with a transaction location of the transaction;
determining, based on the transaction record comprising the merchant category information, a predetermined list of objects associated with the merchant category information;
determining, based on the predetermined list of objects associated with the merchant category information and based on one or more objects in the received video, and using a machine classifier, a likelihood that the user is located at the environment associated with the transaction location by providing a representation of the merchant category information associated with the transaction location and the one or more objects in the received video as inputs to the machine classifier; and receiving, as output from the machine classifier, the likelihood that the user is located in the environment associated with the transaction location;
determining, based on the likelihood that the user is located at the environment associated with the transaction location, that the transaction is not fraudulent; and
based on determining that the transaction is not fraudulent, allowing the transaction to proceed.

2. The method of claim 1 , wherein determining the likelihood that the user is located at the environment associated with the transaction location further comprises:
determining a scene from the received video,
wherein determining the likelihood that the user is located at the environment associated with the transaction location further comprises determining that the scene from the received video is associated with the environment.

3. The method of claim 2 , further comprising:
determining location data associated with the user; and
wherein the scene is determined based on the determined location data associated with the user.

4. The method of claim 2 , wherein the determining the scene from the received video is associated with the environment comprises:
determining, with a scene detection neural network, that the scene from the received video is associated with the environment.

5. The method of claim 1 , further comprising:
identifying the one or more objects from the received video; and
determining whether the identified one or more objects are associated with the environment based on a comparison of the identified one or more objects with the predetermined list of objects,
wherein the determining that the transaction is not fraudulent is further based on a quantity of identified objects that are associated with the environment exceeding a predetermined threshold value.

6. The method of claim 5 , wherein the one or more objects comprise text, the method further comprising:
recognizing the text, and
wherein the determining that the transaction is not fraudulent is further based on the recognizing the text.

7. The method of claim 1 , further comprising:
obtaining permission of the user to access the video feed, and
wherein receiving the video is based on the obtaining permission of the user.

8. The method of claim 1 , further comprising:
sending, to the AR device or a computing device associated with the user, a message requesting that the user capture verification video with the AR device;
receiving verification video from the AR device; and
determining, based on the verification video, that the transaction is not fraudulent.

9. The method of claim 8 , wherein determining, from the verification video, that the transaction is not fraudulent further comprises:
determining, from the verification video, that that the user performed a verification action; and
determining that the transaction is not fraudulent based on determining that the user performed the verification action.

10. The method of claim 1 , wherein the video feed comprises a historical video feed.

11. A fraud detection server system comprising:
one or more processors; and
a memory storing instructions thereon that, when executed by the one or more processors, cause the fraud detection server system to:
obtain permission to access a video feed from an augmented reality (AR) device associated with a user;
identify a transaction associated with the user, wherein the transaction is flagged as potentially fraudulent and corresponds to a transaction record comprising merchant category information;
receive video from the video feed of the AR device;
determine, based on the transaction record comprising the merchant category information, an environment associated with a transaction location of the transaction;
determine, based on the transaction record comprising the merchant category information, a predetermined list of objects associated with the merchant category information;
determine, based on the predetermined list of objects associated with the merchant category information and one or more objects in the received video, and using a machine classifier, a likelihood that the user is at the environment associated with the transaction location by providing a representation of the merchant category information associated with the transaction location and the one or more objects in the received video as inputs to the machine classifier; and receiving, as output from the machine classifier, the likelihood that the user is located in the environment associated with the transaction location; and
determine, based on the likelihood that the user is located at the environment associated with the transaction location, whether to cancel the transaction based on the likelihood that the transaction is fraudulent.

12. The fraud detection server system of claim 11 , further comprising instructions that, when executed by the one or more processors, cause the fraud detection server system to:
send a message, to a computing device associated with the user or to the AR device, requesting capture of additional video with the AR device; and
based on sending the message, receive additional video from the AR device;
wherein the fraud detection server system further determines whether the transaction is fraudulent based on the additional video received from the AR device.

13. The fraud detection server system of claim 11 , further comprising instructions that, when executed by the one or more processors, cause the fraud detection server system to:
determine, using the machine classifier, a scene from the received video,
wherein the fraud detection server system further determines the likelihood that the user is located at the environment associated with the transaction location based on the determined scene.

14. The fraud detection server system of claim 11 , further comprising instructions that, when executed by the one or more processors, cause the fraud detection server system to:
receive, from the AR device or from a computing device associated with the user, an indication of the one or more objects in the received video;
wherein the fraud detection server system further determines whether to cancel the transaction based on the indication of the one or more objects in the received video.

15. The fraud detection server system of claim 14 , further comprising instructions that, when executed by the one or more processors, cause the f

Assignees

• Capital One Services Llc

Inventors

• Edwards Joshua
• Loucks Eric
• Learned Jacob

Classifications

• G06Q20/4016 Primary

  involving fraud or risk level assessment in transaction processing · CPC title

• G06Q20/4014

  Identity check for transactions · CPC title

• G06Q20/4015

  using location information · CPC title