SYSTEM AND METHOD FOR DETECTION AND TRACKING OF MOVING OBJECTS

• US 20140085545A1
• Filed: 09/26/2012
• Published: 03/27/2014
• Est. Priority Date: 09/26/2012
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Tu, Jilin, Xu, Yi, Del Amo, Ana Isabel, Sebastian, Thomas Baby
• Granted Patent: US 9,465,997 B2
• Time in Patent Office: Days
• US Class Current: 348/659
• CPC Class Codes: G06F 18/254 of classification results, e.g. of results related to same input data
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30236 Traffic on road, railway or crossing
• CPC Class Codes: G06T 7/187 involving region growing; involving region merging; involving connected component labelling
• CPC Class Codes: G06T 7/254 involving subtraction of images
• CPC Class Codes: G06V 10/809 of classification results, e.g. where the classifiers operate on the same input data
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G06V 40/23 Recognition of whole body movements, e.g. for sport training

Abstract

A method implemented using a processor based device is disclosed. The method includes receiving a video stream comprising a plurality of image frames having at least one moving object, determining a difference between at least two image frames among the plurality of image frames and generating a difference image comprising a plurality of image blobs corresponding to the at least one moving object. The method further includes generating a plurality of bounding boxes, each bounding box surrounding at least one corresponding image blob among the plurality of image blobs, and determining a subset of bounding boxes among the plurality of bounding boxes, associated with the corresponding moving object, using a fuzzy technique based on a perceptual characterization of the subset of bounding boxes. The method also includes merging the subset of bounding boxes to generate a merged bounding box enclosing the subset of bounding boxes to detect the moving object.

Claims

1. A method implemented using a processor based device, comprising:

receiving a video stream comprising a plurality of image frames having at least one moving object;

determining a difference between at least two image frames among the plurality of image frames and generating a difference image comprising a plurality of image blobs corresponding to the at least one moving object;

generating a plurality of bounding boxes, each bounding box surrounding at least one corresponding image blob among the plurality of image blobs;

determining a subset of bounding boxes among the plurality of bounding boxes, associated with the corresponding moving object, using a fuzzy technique based on a perceptual characterization of the subset of bounding boxes; and

merging the subset of bounding boxes to generate a merged bounding box enclosing the subset of bounding boxes to detect the moving object.

2. The method of claim 1, wherein the fuzzy technique comprises:

determining a characterizing parameter associated with a pair of bounding boxes among the plurality of bounding boxes based on at-least one of geometrical, motion, and appearance properties of the pair of bounding boxes;

determining a fuzzy parameter associated with the pair of bounding boxes based on the characterizing parameter;

determining a box merging parameter associated with the pair of bounding boxes based on the fuzzy parameter; and

determining a fuzzy distance between the pair of bounding boxes based on the box merging parameter.

3. The method of claim 2, wherein the geometrical property comprises a geometrical affinity of the pair of bounding boxes.

4. The method of claim 2, wherein the motion property comprises a motion cohesion between the pair of bounding boxes.

5. The method of claim 2, wherein the appearance property comprises an appearance similarity between the pair of bounding boxes.

6. The method of claim 2, wherein the fuzzy parameter comprises a linguistic variable determined based on the characterizing parameter and a membership function.

7. The method of claim 6, wherein the membership function comprises a gaussian function, or a sigmoid function.

8. The method of claim 2, wherein determining the box merging parameter comprises:

determining the geometrical, the motion, and the appearance properties comprising a geometrical affinity, a motion cohesion and an appearance similarity respectively associated with the pair of bounding boxes;

determining a plurality of fuzzy parameters associated with the pair of bounding boxes, wherein each of the plurality of fuzzy parameters corresponds to one of the geometrical affinity, the motion cohesion and the appearance similarity of the pair of bounding boxes; and

determining a linguistic variable based on a decision rule formulated based on the plurality of fuzzy parameters.

9. The method of claim 1, wherein merging the subset of bounding boxes is based on an agglomerative clustering algorithm.

10. The method of claim 9, wherein merging the subset of bounding boxes comprises merging a pair of bounding boxes to generate the merged bounding box enclosing the pair of bounding boxes.

11. The method of claim 10, further comprising determining an area of the merged bounding box, wherein merging the pair of bounding boxes is based on the determined area of the merged bounding box.

12. The method of claim 10, wherein the perceptual characterization is based on at least one of the geometrical, the motion, and the appearance properties of the pair of bounding boxes.

13. A system comprising:

a processor based device configured to;

receive from a video camera a video stream comprising a plurality of image frames having at least one moving object;

determine a difference between at least two image frames among the plurality of image frames to generate a difference image comprising a plurality of image blobs;

generate a plurality of bounding boxes, each bounding box surrounding at least one corresponding image blob among the plurality of image blobs;

determine a subset of bounding boxes among the plurality of bounding boxes, associated with the corresponding moving object, using a fuzzy technique based on a perceptual characterization of the subset of bounding boxes; and

merge the subset of bounding boxes to generate a merged bounding box enclosing the subset of bounding boxes to detect the moving object.

14. The system of claim 13, wherein the processor based device is configured to use the fuzzy technique comprising:

determining a characterizing parameter associated with a pair of bounding boxes among the plurality of bounding boxes based on at least one of geometrical, motion and appearance properties of the pair of bounding boxes;

determining a fuzzy parameter associated with the pair of bounding boxes based on the characterizing parameter;

determining a box merging parameter associated with the pair of bounding boxes based on the fuzzy parameter; and

determining a fuzzy distance between the pair of bounding boxes based on the box merging parameter.

15. The system of claim 14, wherein the processor based device is further configured to determine at least one of the geometrical, the motion and the appearance properties comprising a box affinity, a motion cohesion and an appearance similarity respectively.

16. The system of claim 14, wherein the processor based device is further configured to determine the fuzzy parameter comprising a linguistic variable determined based on the characterizing parameter and a membership function.

17. The system of claim 14, wherein the processor based device is further configured to determine the box merging parameter by:

determining the geometrical, the motion, and the appearance properties comprising a geometrical affinity, a motion cohesion and an appearance similarity respectively associated with the pair of bounding boxes;

determining a plurality of fuzzy parameters associated with the pair of bounding boxes, wherein each of the plurality of fuzzy parameters corresponds to one of the geometrical affinity, the motion cohesion and the appearance similarity of the pair of bounding boxes; and

determining a linguistic variable based on a decision rule formulated based on the plurality of fuzzy parameters.

18. The system of claim 13, wherein the processor based device is configured to merge a pair of bounding boxes among the subset of bounding boxes to generate a merged bounding box enclosing the pair of bounding boxes.

19. The system of claim 18, wherein the processor based device is configured to determine an area of the merged bounding box, wherein the pair of bounding boxes is merged based on the determined area of the merged bounding box.

20. The system of claim 18, wherein the processor based device is configured to determine the perceptual characterization based on at-least one of the geometrical, the motion, and the appearance properties of the pair of bounding boxes.

21. A non-transitory computer readable medium encoded with a program to instruct a processor based device to:

receive a video stream comprising a plurality of image frames having at least one moving object;

determine a difference between at least two image frames among the plurality of image frames to generate a difference image comprising a plurality of image blobs corresponding to the at least one moving object;

generate a plurality of bounding boxes, each bounding box surrounding at least one corresponding image blob among the plurality of image blobs;

determine a subset of bounding boxes among the plurality of bounding boxes, associated with the corresponding moving object, using a fuzzy technique based on a perceptual characterization of the subset of bounding boxes; and

merge the subset of bounding boxes to generate a merged bounding box enclosing the subset of bounding boxes to detect the moving object.