Granted Patent B2

US 12,323,459 · App. 17/957,775 · Granted Jun 3, 2025

Inventors: Björn Völcker (Lund, SE); Stefan Lundberg (Lund, SE)

Assignee: AXIS AB

H04L63/1466G08B13/19634G08B13/19667

Abstract

Claims (58)

1. A method of detecting a video replay attack where a camera captures images of video played on a display screen instead of a live scene, the method comprising:

receiving image data at a first bit-depth, the image data having a first intensity range;

mapping the image data, using a mapping operation, from the first bit-depth to a second, reduced bit-depth, to produce a representation of the image data having a second intensity range;

generating metadata characterizing the mapping operation;

associating the metadata with the representation of the image data;

determining the first intensity range from the representation of the image data;

determining the second intensity range from the representation of the image data;

determining an expected mapping operation on basis of the second intensity range and the first intensity range;

comparing the mapping operation and the expected mapping operation; and

determining that the image data results from a video replay attack if a measure of the difference between the mapping operation and the expected mapping operation exceeds a threshold level.

2. The method of claim 1 , wherein associating the metadata with the representation of the image data is performed by encoding the metadata and the representation of the image data into an encoded video comprising one or more image frames.

3. The method of claim 2 , wherein the metadata is encoded in a header of one or more encoded image frames in the encoded video.

4. The method of claim 2 , further comprising digitally signing the encoded video.

5. The method of claim 1 , where the step of comparing the mapping operations comprises comparing a diversity measure of image pixel change amounts in the mapping operation.

6. The method of claim 5 , wherein the diversity measure is based on the number of uniquely different pixel change amounts in the mapping operation.

7. The method of claim 1 , wherein the representation of the image data comprises information specifying the first intensity range.

8. The method of claim 7 , wherein the information specifying the first intensity range is stored in a header of an image, a header of a group of images or a header of a video file.

9. The method of claim 1 , wherein the first bit-depth is at least 10, and the second, reduced bit-depth is at most 9.

10. A non-transitory computer readable storage medium having stored thereon instructions for implementing a method for detecting a video replay attack, when executed on a device having processing capabilities, the method comprising:

receiving image data at a first bit-depth, the image data having a first intensity range;

mapping the image data, using a mapping operation, from the first bit-depth to a second, reduced bit-depth, to produce a representation of the image data having a second intensity range;

generating metadata characterizing the mapping operation;

associating the metadata with the representation of the image data;

determining the first intensity range from the representation of the image data;

determining the second intensity range from the representation of the image data;

determining an expected mapping operation on basis of the second intensity range and the first intensity range;

comparing the mapping operation and the expected mapping operation; and

determining that the image data results from a video replay attack if a measure of the difference between the mapping operation and the expected mapping operation exceeds a threshold level.

11. A method of detecting a video replay attack where a camera is capturing images of video played on a display screen instead of a live scene, the method comprising:

receiving a representation of image data having a second intensity range, and metadata characterizing a mapping operation that was used to map the image data having a first intensity range from a first bit-depth to a second, reduced bit-depth, thereby generating the representation of image data having the second intensity range;

determining the first intensity range from the representation of the image data;

determining the second intensity range from the representation of the image data;

determining an expected mapping operation on basis of the second intensity range and the first intensity range;

comparing the mapping operation and the expected mapping operation; and

determining that the image data results from a video replay attack if a measure of the difference between the mapping operation and the expected mapping operation exceeds a threshold level.

12. A system for detecting a video replay attack where a camera is capturing images of video played on a display screen instead of a live scene, the system comprising:

a memory;

a processor, coupled to the memory, configured to:

receive image data at a first bit-depth, the image data having a first intensity range;

map the image data, using a mapping operation, from the first bit-depth to a second, reduced bit-depth, to produce a representation of the image data having a second intensity range;

generate metadata characterizing the mapping operation;

associate the metadata with the representation of the image data;

determine the first intensity range from the representation of the image data;

determine the second intensity range from the representation of the image data;

determine an expected mapping operation on basis of the second intensity range and the first intensity range;

compare the mapping operation and the expected mapping operation; and

determine that the image data results from a video replay attack if a measure of the difference between the mapping operation and the expected mapping operation exceeds a threshold level.

13. The system of claim 12 , wherein the association of the metadata with the representation of the image data is performed by encoding the metadata and the representation of the image data into an encoded video comprising one or more image frames.

14. The system of claim 12 , wherein the processor is further configured to transmit video to a client.

15. A video client configured for detecting a video replay attack where a camera is capturing images of video played on a display screen instead of a live scene, the client comprising:

a memory;

a processor, coupled to the memory, configured to:

receive a representation of image data having a second intensity range and metadata characterizing a mapping operation that was used to map the image data having a first intensity range from a first bit-depth to a second, reduced bit-depth, thereby generating the representation of image data having the second intensity range;

determine the first intensity range from the representation of the image data;

determine the second intensity range from the representation of the image data;

determine an expected mapping operation on basis of the second intensity range and the first intensity range;

compare the mapping operation and the expected mapping operation; and

determine that the image data results from a video replay attack if a measure of the difference between the mapping operation and the expected mapping operation exceeds a threshold level.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (33)