Secure human user verification for electronic systems

Patent metadata

Field	Value
Publication number	US-12380198-B2
Application number	US-202318352739-A
Country	US
Kind code	B2
Filing date	Jul 14, 2023
Priority date	Jul 19, 2022
Publication date	Aug 5, 2025
Grant date	Aug 5, 2025

Abstract

Systems and methods are provided for a secure human user verification procedure that involves physical manipulation of a user device with respect to its environment. A challenge prompt may be presented for a user to provide challenge response data. A subsystem of an application running on the user device may be configured (e.g., using a machine learning model) to evaluate the challenge response data and determine whether the challenge response data includes the subject of the challenge. To ensure that the user is not providing pre-existing data in response to the challenge prompt, data items associated with the challenge response data may be evaluated to validate the challenge response data as being provided by a human user rather than automatically generated.

First claim

What is claimed is:

1. A computing system comprising:
a display device;
a camera device configured to generate image data;
computer-readable memory storing an image recognition model configured to recognize images of a plurality of challenge subjects; and
one or more processors programmed by executable instructions to at least:
determine to verify whether a user, of an application configured to be executed by the one or more processors, is a human user;
select a challenge subject from the plurality of challenge subjects, wherein an image of the challenge subject is to be provided in response to a challenge prompt;
present the challenge prompt using the display device;
receive, in response to the challenge prompt, the image data and challenge context data;
determine, using the image recognition model, that at least a portion of the image data represents the image;
determine that the challenge context data fails to satisfy a response validation criterion based on determining that the challenge context data indicates the image data was generated using another camera device different from the camera device, wherein the challenge context data represents a source of the image data; and
in response to determining that the challenge context data fails to satisfy the response validation criterion, prevent an operation of the application.

2. The computing system of claim 1 , wherein the image recognition model comprises a neural network trained to recognize the plurality of challenge subjects.

3. The computing system of claim 1 , wherein the plurality of challenge subjects comprise objects expected to be within an environment of the user.

4. The computing system of claim 1 , wherein the image data and challenge context data are not transmitted to a second computing system.

5. The computing system of claim 1 , further comprising a movement sensor, wherein to determine that the challenge context data fails to satisfy the response validation criterion, the one or more processors are programmed by the executable instructions to determine that movement detected by the movement sensor is not indicative of human use of the camera device to generate the image data, wherein the challenge context data represents the movement detected by the movement sensor.

6. The computing system of claim 1 , wherein to determine that the challenge context data fails to satisfy the response validation criterion, the one or more processors are programmed by the executable instructions to determine that the image data was generated at a first geographic location different from a second geographic location, wherein the challenge context data represents the first geographic location.

7. The computing system of claim 1 , wherein to determine that the challenge context data fails to satisfy the response validation criterion, the one or more processors are programmed by the executable instructions to determine that the image data was generated prior to presentation of the challenge prompt, wherein the challenge context data represents a time at which the image data was generated.

8. The computing system of claim 1 , wherein to determine that the challenge context data fails to satisfy the response validation criterion, the one or more processors are programmed by the executable instructions to determine that the image data was generated less than a threshold period of time after presentation of the challenge prompt, wherein the challenge context data represents a time at which the image data was generated.

9. The computing system of claim 1 , wherein to determine that the challenge context data fails to satisfy the response validation criterion, the one or more processors are programmed by the executable instructions to determine that the image data was generated more than a threshold period of time after presentation of the challenge prompt, wherein the challenge context data represents a time at which the image data was generated.

10. A computer-implemented method comprising:
under control of a computing system comprising one or more processors configured to execute specific instructions:
determining that a human user verification procedure is to be performed;
selecting a challenge subject from a plurality of challenge subjects, wherein an image of the challenge subject is to be provided in response to a challenge prompt;
presenting the challenge prompt using a display device of the computing system;
receiving, in response to the challenge prompt, image data and challenge context data;
generating subject verification data representing whether the image data comprises a representation of the image;
generating response validation data representing whether the challenge context data satisfies a response validation criterion based on determining that the image data was generated prior to presentation of the challenge prompt, wherein the challenge context data represents a time at which the image data was generated; and
generating human user verification data based at least partly on the subject verification data and the response validation data.

11. The computer-implemented method of claim 10 , wherein determining that the human user verification procedure is to be performed is based on a request received from a server computing device over a network, and wherein neither the image data nor the challenge context data are transmitted by the computing system to the server computing device.

12. The computer-implemented method of claim 10 , further comprising determining that the challenge context data fails to satisfy the response validation criterion, wherein the human user verification data represents a failure to verify that a human user responded to the challenge prompt.

13. The computer-implemented method of claim 12 , wherein determining that the challenge context data fails to satisfy the response validation criterion comprises determining that the image data was generated using a camera device external to the computing system, and wherein the challenge context data represents a source of the image data.

14. The computer-implemented method of claim 12 , wherein determining that the challenge context data fails to satisfy the response validation criterion comprises determining that movement of the computing system is not indicative of human use of a camera device of the computing system to generate the image data, wherein the challenge context data represents movement of the computing system.

15. The computer-implemented method of claim 12 , wherein determining that the challenge context data fails to satisfy the response validation criterion comprises determining that the image data was generated at first geographic location different from a second geographic location, wherein the challenge context data represents the first geographic location.

16. The computer-implemented method of claim 12 , wherein determining that the challenge context data fails to satisfy the response validation criterion comprises determining that the image data was generated less than a threshold period of time after presentation of the challenge prompt, wherein the challenge context data represents a time at which the image data was generated.

17. The computer-implemented method of claim 12 , further comprising, in response to determining that the challenge context data fails to satisfy the response validation criterion, quarantining data associated with a non-genuine human user.

18. The computer-implemented method of claim 12 , further comprising, in response to determining that the challenge context data fai

Assignees

• Tealium Inc

Inventors

• Rouse Craig P

Classifications

• G06V10/255

  Detecting or recognising potential candidate objects based on visual cues, e.g. shapes · CPC title

• H04L9/3271

  using challenge-response · CPC title

• G06F2221/2133

  Verifying human interaction, e.g., Captcha · CPC title

• G06F21/36 Primary

  by graphic or iconic representation · CPC title