Classifying images and videos

Patent metadata

Field	Value
Publication number	US-10657378-B2
Application number	US-201615754249-A
Country	US
Kind code	B2
Filing date	Aug 31, 2016
Priority date	Sep 25, 2015
Publication date	May 19, 2020
Grant date	May 19, 2020

Abstract

A method, system and computer program product for classifying an image or video. An image or video to be classified is received. Scene statistics (statistical model of pictures, images or videos representative of pictures, images or videos, respectively, that are captured of the physical world) of the image or video are captured. A model (a statistical model that describes a set of probability distributions) of the image or video is then created using the captured scene statistics. A comparison between the model of the image or video with two other models of images or videos is performed, such as a model of visible light images or videos and a model of infrared images or videos. The received image or video is then classified (e.g., classified as corresponding to a visible light image) based on the comparison.

First claim

The invention claimed is:

1. A method for classifying an image or video, the method comprising:
receiving an image or video to be classified;
capturing scene statistics in said image or video;
creating a model of said image or video using said captured scene statistics;
comparing, by a processor, said model of said image or video with a model of visible light images or videos and with a model of infrared images or videos; and
classifying, by said processor, said image or video as a visible light or video or an infrared light or video based on said comparison.

2. The method as recited in claim 1 further comprising:
capturing scene statistics in visible light images or videos to create said model of visible light images or videos.

3. The method as recited in claim 1 further comprising:
capturing scene statistics in infrared images or videos to create said model of infrared images or videos.

4. The method as recited in claim 1 , wherein said comparison determines how similar is a probability distribution of said model of said image or video with a probability distribution of said model of visible light images or videos and with a probability distribution of said model of infrared images or videos.

5. The method as recited in claim 4 , wherein a similarity between said probability distribution of said model of said image or video with said probability distribution of said model of visible light images or videos and with said probability distribution of said model of infrared images or videos is determined by computing one of the following: Mahalanobis distance, Eucledian distance, mean-square error, absolute error, earth mover distance, city block distance, Minkowski distance, Sorensen distance, Gower distance, Soergel distance, Kulczynski distance, Canberra distance, Lorentzian distance and Chebyshev distance.

6. The method as recited in claim 4 , wherein a similarity between said probability distribution of said model of said image or video with said probability distribution of said model of visible light images or videos and with said probability distribution of said model of infrared images or videos is determined using a machine-learning model.

7. A computer program product for classifying an image or video, the computer program product comprising a computer readable storage medium having program code embodied therewith, the program code comprising the programming instructions for:
receiving an image or video to be classified;
capturing scene statistics in said image or video;
creating a model of said image or video using said captured scene statistics;
comparing said model of said image or video with a model of visible light images or videos and with a model of infrared images or videos; and
classifying said image or video as a visible light or video or an infrared light or video based on said comparison.

8. The computer program product as recited in claim 7 , wherein the program code further comprises the programming instructions for:
capturing scene statistics in visible light images or videos to create said model of visible light images or videos.

9. The computer program product as recited in claim 7 , wherein the program code further comprises the programming instructions for:
capturing scene statistics in infrared images or videos to create said model of infrared images or videos.

10. The computer program product as recited in claim 7 , wherein said comparison determines how similar is a probability distribution of said model of said image or video with a probability distribution of said model of visible light images or videos and with a probability distribution of said model of infrared images or videos.

11. The computer program product as recited in claim 10 , wherein a similarity between said probability distribution of said model of said image or video with said probability distribution of said model of visible light images or videos and with said probability distribution of said model of infrared images or videos is determined by computing one of the following: Mahalanobis distance, Eucledian distance, mean-square error, absolute error, earth mover distance, city block distance, Minkowski distance, Sorensen distance, Gower distance, Soergel distance, Kulczynski distance, Canberra distance, Lorentzian distance and Chebyshev distance.

12. The computer program product as recited in claim 10 , wherein a similarity between said probability distribution of said model of said image or video with said probability distribution of said model of visible light images or videos and with said probability distribution of said model of infrared images or videos is determined using a machine-learning model.

13. A system, comprising:
a memory unit for storing a computer program for classifying an image or video; and
a processor coupled to the memory unit, wherein the processor is configured to execute the program instructions of the computer program comprising:
receiving an image or video to be classified;
capturing scene statistics in said image or video;
creating a model of said image or video using said captured scene statistics;
comparing said model of said image or video with a model of visible light images or videos and with a model of infrared images or videos; and
classifying said image or video as a visible light or video or an infrared light or video based on said comparison.

14. The system as recited in claim 13 , wherein the program instructions of the computer program further comprise:
capturing scene statistics in visible light images or videos to create said model of visible light images or videos.

15. The system as recited in claim 13 , wherein the program instructions of the computer program further comprise:
capturing scene statistics in infrared images or videos to create said model of infrared images or videos.

16. The system as recited in claim 13 , wherein said comparison determines how similar is a probability distribution of said model of said image or video with a probability distribution of said model of visible light images or videos and with a probability distribution of said model of infrared images or videos.

17. The system as recited in claim 16 , wherein a similarity between said probability distribution of said model of said image or video with said probability distribution of said model of visible light images or videos and with said probability distribution of said model of infrared images or videos is determined by computing one of the following: Mahalanobis distance, Eucledian distance, mean-square error, absolute error, earth mover distance, city block distance, Minkowski distance, Sorensen distance, Gower distance, Soergel distance, Kulczynski distance, Canberra distance, Lorentzian distance and Chebyshev distance.

18. The system as recited in claim 16 , wherein a similarity between said probability distribution of said model of said image or video with said probability distribution of said model of visible light images or videos and with said probability distribution of said model of infrared images or videos is determined using a machine-learning model.

19. A method for classifying an image or video, the method comprising:
receiving an image or video to be classified;
capturing scene statistics in said image or video;
creating a model of said image or video using said captured scene statistics;
comparing, by a processor, said model of said image or video with a first model of images or videos and with a second model of images or videos; and
classifying, by said proce

Assignees

• Univ Texas

Inventors

• Bovik Alan
• Goodall Todd

Classifications

• G06K9/6212

  Physics · mapped topic

• G06K9/00718 Primary

  Physics · mapped topic

• G06K9/6277

  Physics · mapped topic

• G06K9/00771

  Physics · mapped topic

• G06K9/00744

  Physics · mapped topic