Methods for detecting and recognizing a moving object in video and devices thereof

• US 9,288,450 B2
• Filed: 08/09/2012
• Issued: 03/15/2016
• Est. Priority Date: 08/18/2011
• Status: Active Grant

• Current Assignee: Infosys Limited
• Original Assignee: Infosys Limited
• Inventors: Rai, Harikrishna G. N., Hota, Rudra Narayana, Jonna, Kishore, Krishna, P. Radha
• Primary Examiner(s): Bitar, Nancy
• Application Number:
  US13/570,350
• Publication Number:
  US 20130044914A1
• Time in Patent Office: 1,314 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06V 10/245 by locating a pattern; Special marks for positioning
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources

Abstract

A method, non-transitory computer readable medium, and apparatus that extracts at least one key image from one or more images of an object. Outer boundary markers for an identifier of the object in the at least one key image are detected. An identification sequence from the identifier of the object between the outer boundary markers in the at least one key image is recognized. The recognized identification sequence of the object in the at least one key image is provided.

Claims

1. A method for detecting and recognizing a moving object in video, the method comprising:

extracting, by a tracking management computing device, at least one key image from one or more images of an object;

detecting, by the tracking management computing device, one or more potential boundary markers for an identifier of the object in the at least one key image;

determining, by the tracking management computing device, a size of each of the one or more potential boundary markers based on color segmentation of the at least one key image;

filtering out, by the tracking management computing device, each of the one or more potential boundary markers with a determined size that is outside a threshold size range, wherein the one or more potential boundary markers that were not filtered out comprise outer boundary markers;

recognizing, by the tracking management computing device, an identification sequence from the identifier of the object between the outer boundary markers in the at least one key image; and

providing, by the tracking management computing device, the recognized identification sequence of the object in the at least one key image.

2. The method as set forth in claim 1, wherein the detecting the one or more potential boundary markers further comprises:

filtering out, by the tracking management computing device, the one or more potential boundary markers in the at least one key image based on a distance between the one or more potential boundary markers.

3. The method as set forth in claim 1, wherein the detecting further comprises:

adjusting, by the tracking management computing device, an orientation of the one or more potential boundary markers in the at least one key image with respect to a fixed axis.

4. The method as set forth in claim 3, wherein the adjusting further comprises:

generating, by the tracking management computing device, a bounding box around the one or more potential boundary markers;

computing, by the tracking management computing device, the orientation of the generated bounding box with respect to the fixed axis; and

rotating, by the tracking management computing device, the at least one key image based on the computed orientation.

5. The method as set forth in claim 1, wherein the identifying further comprises:

cropping, by the tracking management computing device, the outer boundary markers from a region in the at least one key image before the identifying.

6. The method as set forth in claim 5, wherein the identifying further comprises:

thresholding, by the tracking management computing device, the cropped region to identify one or more white areas and one or more black areas.

7. The method as set forth in claim 6, wherein the identifying further comprises

assigning, by the tracking management computing device, based on a sliding window, one value to the one or more identified white areas and another value to the one or more identified black areas.

8. The method as set forth in claim 7, wherein the identifying further comprises:

determining, by the traffic management computing device, when the identified identification sequence from the identifier of the object between the outer boundary markers in the at least one key image is correct, wherein the cropped region is rotated 180 degrees and the identifying is repeated when the determined identification sequence is incorrect.

9. The method as set forth in claim 8, wherein the identifying further comprises:

converting, by the tracking management computing device, the identification sequence into a decimal number for the providing the identified identification sequence.

29. The method as set forth in claim 4, wherein the bounding box is generated based on spatial pixel manipulation of the one or more potential boundary markers.

10. A non-transitory computer readable medium having stored thereon instructions for detecting and recognizing a moving object in video comprising machine executable code which when executed by at least one processor, causes the processor to perform steps comprising:

extracting at least one key image from one or more images of an object;

detecting one or more potential boundary markers for an identifier of the object in the at least one key image;

determining a size of each of the one or more potential boundary markers based on color segmentation of the at least one key image;

filtering out each of the one or more potential boundary markers with a determined size that is outside a threshold size range, wherein the one or more potential boundary markers that were not filtered out comprise outer boundary markers;

recognizing an identification sequence from the identifier of the object between the outer boundary markers in the at least one key image; and

providing the recognized identification sequence of the object in the at least one key image.

11. The medium as set forth in claim 10, wherein the detecting the one or more potential boundary markers further comprises:

filtering the one or more potential boundary markers in the at least one key image based on a distance between the one or more potential boundary markers.

12. The medium as set forth in claim 10, wherein the detecting further comprises:

adjusting an orientation of the one or more potential boundary markers in the at least one key image with respect to a fixed axis.

13. The medium as set forth in claim 12, wherein the adjusting further comprises:

generating a bounding box around the one or more potential boundary markers;

computing the orientation of the generated bounding box with respect to the fixed axis; and

rotating the at least one key image based on the computed orientation.

14. The medium as set forth in claim 10, wherein the identifying further comprises:

cropping the outer boundary markers from a region in the at least one key image before the identifying.

15. The medium as set forth in claim 14, wherein the identifying further comprises:

thresholding the cropped region to identify one or more white areas and one or more black areas.

16. The medium as set forth in claim 15, wherein the identifying further comprises:

assigning, based on a sliding window, one value to the one or more identified white areas and another value to the one or more identified black areas.

17. The medium as set forth in claim 16, wherein the identifying further comprises:

determining when the identified identification sequence from the identifier of the object between the outer boundary markers in the at least one key image is correct, wherein the cropped region is rotated 180 degrees and the identifying is repeated when the determined identification sequence is incorrect.

18. The medium as set forth in claim 17, wherein the identifying further comprises:

converting the identification sequence into a decimal number for the providing the identified identification sequence.

30. The medium as set forth in claim 13, wherein the bounding box is generated based on spatial pixel manipulation of the one or more potential boundary markers.

19. A tracking management computing device comprising:

one or more processors;

a memory coupled to the one or more processors which are configured to be capable of executing programmed instructions comprising and stored in the memory to;

extract at least one key image from one or more images of an object;

detect one or more potential boundary markers for an identifier of the object in the at least one key image;

determine a size of each of the one or more potential boundary markers based on color segmentation of the at least one key image;

filter out each of the one or more potential boundary markers with a determined size that is outside a threshold size range, wherein the one or more potential boundary markers that were not filtered out comprise outer boundary markers;

recognize an identification sequence from the identifier of the object between the outer boundary markers in the at least one key image; and

provide the recognized identification sequence of the object in the at least one key image.

20. The device as set forth in claim 19, wherein the one or more processors is further configured to be capable of executing programmed instructions for the detecting the outer boundary markers comprising and stored in the memory to:

filter the one or more potential boundary markers in the at least one key image based on a distance between the one or more potential boundary markers.

21. The device as set forth in claim 19, wherein the one or more processors is further configured to be capable of executing programmed instructions for the detecting comprising and stored in the memory to:

adjust an orientation of the one or more potential boundary markers in the at least one key image with respect to a fixed axis.

22. The device as set forth in claim 21, wherein the one or more processors is further configured to be capable of executing programmed instructions for the adjusting comprising and stored in the memory to:

generate a bounding box around the one or more potential boundary markers;

computer the orientation of the generated bounding box with respect to the fixed axis; and

rotating the at least one key image based on the computed orientation.

23. The device as set forth in claim 19, wherein the one or more processors is further configured to be capable of executing programmed instructions for the identifying comprising and stored in the memory to:

crop the outer boundary markers from a region in the at least one key image before the identifying.

24. The device as set forth in claim 23, wherein the one or more processors is further configured to be capable of executing programmed instructions for the identifying comprising and stored in the memory to:

threshold the cropped region to identify one or more white areas and one or more black areas.

25. The device as set forth in claim 24, wherein the one or more processors is further configured to be capable of executing programmed instructions for the identifying further comprising and stored in the memory to:

assign, based on a sliding window, one value to the one or more identified white areas and another value to the one or more identified black areas.

26. The device as set forth in claim 25, wherein the one or more processors is further configured to be capable of executing programmed instructions for the identifying comprising and stored in the memory to:

determine when the identified identification sequence from the identifier of the object between the outer boundary markers in the at least one key image is correct, wherein the cropped region is rotated 180 degrees and the identifying is repeated when the determined identification sequence is incorrect.

27. The device as set forth in claim 26, wherein the one or more processors is further configured to be capable of executing programmed instructions stored in the memory for the identifying comprising and stored in the memory to:

convert the identification sequence into a decimal number for the providing the identified identification sequence.

28. The device as set forth in claim 22, wherein the bounding box is generated based on spatial pixel manipulation of the one or more potential boundary markers.