Video fingerprint system and method thereof

• US 10,339,660 B2
• Filed: 07/01/2016
• Issued: 07/02/2019
• Est. Priority Date: 05/19/2016
• Status: Active Grant

• Current Assignee: KAMI Vision Incorporated, Xiaoyi Technology Co., Ltd.
• Original Assignee: Shanghai Xiaoyi Technology Co., Ltd.
• Inventors: Wang, Kongqiao
• Primary Examiner(s): Rao, Anand S
• Application Number:
  US15/200,611
• Publication Number:
  US 20170337696A1
• Time in Patent Office: 1,096 Days
• Field of Search: 348160-162
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Abstract

A system used for generating a transformed representation of a quantity of video data structured as a plurality of frames including arrays of rows and columns of pixels having pixel properties. The system may generate first representations of the video data based on a plurality of the rows; generate second representations of the video data based on a plurality of the columns; generate frame representations corresponding to the frames and based on the first and second representations; and combine the frame representations to form the transformed representation of the video data. The system may also generate frame representations respectively corresponding to the frames; combine the frame representations to form a transformed representation of the video data; analyze the transformed representation; and identify frames of interest based on the analysis.

Claims

1. A method for generating a transformed representation of a quantity of video data structured as a plurality of frames including arrays of rows and columns of pixels having pixel properties, comprising:

generating first representations of the video data based on a plurality of the rows;

wherein generating the first representations includes determining first pluralities of statistical values based on pixel property values of pixels in the rows;

generating second representations of the video data based on a plurality of the columns;

wherein generating the second representations includes determining second pluralities of statistical values based on pixel property values of pixels in the columns;

generating frame representations corresponding to the frames and based on the first and second representations; and

combining the frame representations to form the transformed representation of the video data.

2. The method of claim 1, further comprising:

selecting a subset of the rows;

selecting a subset of the columns; and

generating the first and second representations based on the row subsets and column subsets, respectively.

3. The method of claim 1, wherein generating the frame representations further comprises combining the first and second pluralities of statistical values.

4. The method of claim 1, wherein generating the frame representations further comprises:

converting the first and second pluralities of statistical values into video data representing pixels; and

structuring pixel video data into columns of pixels respectively corresponding to the frames.

5. A method of analyzing a quantity of video data structured as a plurality of frames, comprising:

generating frame representations respectively corresponding to the frames;

combining the frame representations to form a transformed representation of the video data;

analyzing the transformed representation; and

identifying frames of interest based on the analysis;

wherein the identified frames of interest include frames with at least one of an activity, a moving event, or a change in a quantity of video data.

6. The method of claim 5, wherein:

the frame representations are structured as video data representing columns of pixels; and

the transformed representation comprises an image.

7. The method of claim 6, wherein analyzing the transformed representation comprises analyzing statistical values of one or more pixel properties of the image.

8. The method of claim 5, wherein the transformed representation comprises a graph.

9. The method of claim 8, wherein analyzing the transformed representation comprises identifying at least one of local maxima of a graph or local minima of the graph.

10. A system for generating a transformed representation of a quantity of video data structured as a plurality of frames including arrays of rows and columns of pixels having pixel properties, comprising:

a memory storing instructions; and

a processor configured to execute the instructions to;

generate first representations of the video data based on a plurality of the rows;

wherein generating the first representations includes determining first pluralities of statistical values based on pixel property values of pixels in the rows;

generate second representations of the video data based on a plurality of the columns;

wherein generating the second representations includes determining second pluralities of statistical values based on pixel property values of pixels in the columns;

generate frame representations corresponding to the frame and based on the first and second representations; and

combine the frame representations to form the transformed representation of the video data.

11. The system of claim 10, wherein generating the frame representations further comprises:

converting the first and second pluralities of statistical values into video data representing pixels; and

structuring pixel video data into columns of pixels respectively corresponding to the frames.

12. A system of analyzing a quantity of video data structured as a plurality of frames, comprising:

a memory storing instructions; and

a processor configured to execute the instructions to;

generate frame representations respectively corresponding to the frames;

combine the frame representations to form a transformed representation of the video data;

analyze the transformed representation;

identify frames of interest based on the analysis;

wherein the identified frames of interest include frames with at least one of an activity, a moving event, or a change in a quantity of video data.

13. The system of claim 12, wherein:

the frame representations are structured as video data representing columns of pixels; and

the transformed representation comprises an image.

14. A non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform a method for generating a transformed representation of a quantity of video data structured as a plurality of frames including arrays of rows and columns of pixels having pixel properties, the method comprising:

generating first representations of the video data based on a plurality of the rows;

wherein generating the first representations includes determining first pluralities of statistical values based on pixel property values of pixels in the rows;

generating second representations of the video data based on a plurality of the columns;

wherein generating the second representations includes determining second pluralities of statistical values based on pixel property values of pixels in the columns;

generating frame representations corresponding to the frames and based on the first and second representations; and

combining the frame representations to form the transformed representation of the video data.

15. The non-transitory computer-readable storage medium of claim 14,

wherein generating the frame representations further comprises:

converting the first and second pluralities of statistical values into video data representing pixels; and

structuring pixel video data into columns of pixels respectively corresponding to the frames.

16. A non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor, cause the at least one processor to perform a method for generating a transformed representation of a quantity of video data structured as a plurality of frames including arrays of rows and columns of pixels having pixel properties, the method comprising:

generating frame representations respectively corresponding to the frames;

combining the frame representations to form a transformed representation of the video data;

analyzing the transformed representation; and

identifying frames of interest based on the analysis;

wherein the identified frames of interest include frames with at least one of an activity, a moving event, or a change in a quantity of video data.

17. The non-transitory computer-readable storage medium of claim 16, wherein:

the frame representations are structured as video data representing columns of pixels; and

the transformed representation comprises an image.