Privacy-preserving image sensor

Patent metadata

Field	Value
Publication number	US-12014558-B2
Application number	US-202217976648-A
Country	US
Kind code	B2
Filing date	Oct 28, 2022
Priority date	Oct 28, 2022
Publication date	Jun 18, 2024
Grant date	Jun 18, 2024

Abstract

In some embodiments, an image sensor is provided. A signal processing unit of the image sensor is configured with executable instructions that cause the signal processing unit to perform actions comprising: reading a captured scene from a pixel array; reading a value from a test initiation register; in response to determining that the value indicates a test mode: processing the captured scene to detect a region of interest associated with the value; and providing the region of interest to a first interface for transmission to a host device; and in response to determining that the value does not indicate the test mode: analyzing the captured scene using a computer vision technique to selectively generate a signal based on the analysis of the captured scene; and selectively providing the signal to a second interface for transmission to the host device.

First claim

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A signal processing unit configured to perform actions comprising:
reading a captured scene from a pixel array;
reading a value from a test initiation register;
in response to determining that the value indicates a test mode:
processing the captured scene to detect a region of interest associated with the value; and
providing the region of interest to a first interface for transmission to a host device; and

in response to determining that the value does not indicate the test mode:
analyzing the captured scene using a computer vision technique to selectively generate a signal based on the analysis of the captured scene; and
selectively providing the signal to a second interface for transmission to the host device.

2. The signal processing unit of claim 1 , wherein processing the captured scene to detect the region of interest includes:
determining a predefined test scene associated with the value;
detecting the predefined test scene in the captured scene; and
cropping the captured scene to a region of interest associated with the predefined test scene.

3. The signal processing unit of claim 1 , wherein selectively generating the signal based on the analysis of the captured scene includes generating the signal based on a detection of a predetermined object in the captured scene.

4. The signal processing unit of claim 1 , wherein analyzing the captured scene using the computer vision technique to selectively generate the signal based on the analysis of the captured scene includes:
assigning at least one label for at least one object detected in the captured scene; and
selectively generating the signal based on the at least one label.

5. The signal processing unit of claim 4 , wherein selectively providing the signal to the second interface for transmission to the host device includes at least one of:
selectively transmitting a trigger signal to the host device through the second interface based on whether a predetermined object is indicated by the at least one label; or
writing a value based on the at least one label to a register to be queried by the host device through the second interface.

6. The signal processing unit of claim 4 , wherein the signal based on the at least one label indicates at least one of a type of the object detected, a gesture of the object detected, a movement of the object detected, or a location of the object detected.

7. The signal processing unit of claim 4 , wherein the at least one object is a human, and wherein the signal based on the labels indicates a mental state of the human.

8. An image sensor, comprising:
a pixel array;
a test initiation register;
a signal processing unit communicatively coupled to the pixel array and the test initiation register; and
one or more interfaces communicatively coupled to the test initiation register and the signal processing unit;
wherein the signal processing unit is configured with executable instructions that, in response to execution by the signal processing unit, cause the signal processing unit to perform actions comprising:
reading a captured scene from the pixel array;
reading a value from the test initiation register;
in response to determining that the value indicates a test mode:
processing the captured scene to detect a region of interest associated with the value; and
providing the region of interest to a first interface of the one or more interfaces for transmission to a host device; and

in response to determining that the value does not indicate the test mode:
analyzing the captured scene using a computer vision technique to selectively generate a signal based on the analysis of the captured scene; and
selectively providing the signal to a second interface of the one or more interfaces for transmission to the host device.

9. The image sensor of claim 8 , wherein processing the captured scene to detect the region of interest includes:
determining a predefined test scene associated with the value;
detecting the predefined test scene in the captured scene; and
cropping the captured scene to a region of interest associated with the predefined test scene.

10. The image sensor of claim 8 , wherein selectively generating the signal based on the analysis of the captured scene includes generating the signal based on a detection of a predetermined object in the captured scene.

11. The image sensor of claim 8 , wherein analyzing the captured scene using the computer vision technique to selectively generate the signal based on the analysis of the captured scene includes:
assigning at least one label for at least one object detected in the captured scene; and
selectively generating the signal based on the at least one label.

12. The image sensor of claim 11 , wherein selectively providing the signal to the second interface of the one or more interfaces for transmission to the host device includes at least one of:
selectively transmitting a trigger signal to the host device through the second interface based on whether a predetermined object is indicated by the at least one label; or
writing a value based on the at least one label to a register to be queried by the host device through the second interface.

13. The image sensor of claim 11 , wherein the signal based on the at least one label indicates at least one of a type of the object detected, a gesture of the object detected, a movement of the object detected, or a location of the object detected.

14. The image sensor of claim 13 , wherein the host device is an intelligent switch, and wherein the type of the object detected indicates whether the object is a human or whether the object is non-human.

15. The image sensor of claim 11 , wherein the host device is an augmented reality system, a virtual reality system, or a mixed reality system, and wherein the object detected is a pupil of an eye.

16. The image sensor of claim 11 , wherein the object detected is a human, and wherein selectively generating the signal based on the at least one label includes generating a signal that indicates a mental state of the human or a biometric characteristic of the human.

17. The image sensor of claim 16 , wherein the host device is a driver monitoring system.

18. A method, comprising:
reading a captured scene from a pixel array;
reading a value from a test initiation register;
in response to determining that the value indicates a test mode:
processing the captured scene to detect a region of interest associated with the value; and
providing the region of interest to a first interface for transmission to a host device; and

in response to determining that the value does not indicate the test mode:
analyzing the captured scene using a computer vision technique to selectively generate a signal based on the analysis of the captured scene; and
selectively providing the signal to a second interface for transmission to the host device.

19. The method of claim 18 , wherein processing the captured scene to detect the region of interest includes:
determining a predefined test scene associated with the value;
detecting the predefined test scene in the captured scene; and
cropping the captured scene to a region of interest associated with the predefined test scene.

20. The method of claim 18 , wherein selectively generating the signal based on the analysis of the captured scene includes generating the signal based on a detection of a

Assignees

• Omnivision Tech Inc

Inventors

• Suess Andreas
• Mu Bo

Classifications

• H04N25/75

  Circuitry for providing, modifying or processing image signals from the pixel array · CPC title

• G06V20/70 Primary

  Labelling scene content, e.g. deriving syntactic or semantic representations · CPC title

• G06V10/25

  Determination of region of interest [ROI] or a volume of interest [VOI] · CPC title

• G06F21/6245

  Protecting personal data, e.g. for financial or medical purposes · CPC title

• H04N5/765

  Interface circuits between an apparatus for recording and another apparatus · CPC title