Hybrid method and system of video and vision based access control for parking stall occupancy determination

• US 9,224,062 B2
• Filed: 08/09/2013
• Issued: 12/29/2015
• Est. Priority Date: 08/09/2013
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Wu, Wencheng, Wang, Yao Rong, Pennington, Graham
• Primary Examiner(s): Liew, Alex
• Application Number:
  US13/963,200
• Publication Number:
  US 20150043771A1
• Time in Patent Office: 872 Days
• Field of Search: 382/103
• US Class Current: 1/1
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06F 18/217 Validation; Performance evaluation; Active pattern learning techniques
• CPC Class Codes: G06F 18/24 Classification techniques
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/20081 Training; Learning
• 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: G06V 20/40 in video content extracting overlay text G06V20/62; video retrieval G06F16/70; processing of video elementary streams in video servers H04N21/234; processing of video elementary streams in video clients H04N21/44
• 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 20/54 of traffic, e.g. cars on the road, trains or boats
• CPC Class Codes: G08G 1/14 indicating individual free spaces in parking areas

Abstract

Hybrid methods, systems and processor-readable media for video and vision based access control for parking occupancy determination. One or more image frames of a parking area of interest can be acquired from among two or more regions of interest defined with respect to the parking area of interest. The regions of interest can be analyzed for motion detection or image content change detection. An image content classification operation can be performed with respect to a first region of interest among the regions of interest based on the result of the image content change detection. An object tracking operation can then be performed with respect to a second region of interest among the regions of interest if the result of the image content classification operation indicates a presence of one or more objects of interest within the parking area of interest.

Claims

1. A hybrid method of video and vision based access control for parking occupancy determination, said method comprising:

acquiring at least one image frame of a parking area of interest from among at least two regions of interest defined with respect to said parking area of interest;

analyzing said at least two regions of interest defined within said parking area of interest for motion detection or content change detection;

performing a vision-based image content classification operation with respect to a first region of interest among said at least two regions of interest based on a result of said content change detection; and

performing an object tracking operation with respect to a second region of interest among said at least two regions of interest if a result of said vision-based image content classification operation indicates a presence of an object of interest within said parking area of interest to thereafter determine a type of access authorized with respect to said parking area of interest and based on a result of said object tracking operation.

2. The method of claim 1 wherein said parking area of interest comprises at least one of a parking lot entrance or a parking lot exit.

3. The method of claim 1 further comprising defining said at least two regions of interest with respect to said parking area of interest.

4. The method of claim 3 further comprising training said vision-based image content classification with respect to said first region of interest for use in performing said image content classification operation.

5. The method of claim 3 wherein said vision-based image content classification includes classification with respect to a presence of at least one vehicle class and without a presence of any vehicle class.

6. The method of claim 5 further comprising training a classifier for performing said image content classification by:

analyzing at least two frames of training videos at said first region of interest for said motion detection;

automatically collect training samples, an image of said first region of interest in one of said at least two frames, if said motion detection exceeds a threshold; and

grouping collected samples by human operation into two classes comprising said presence of said at least vehicle class and said without said presence of any vehicle class.

7. The method of claim 6 further comprising:

smoothing a resulting tracked trajectory with respect to said object tracking operation to generate a smoothed trajectory; and

determining said type of access based on a direction and a travel distance of said smoothed trajectory.

8. A hybrid system of video and vision based access control for parking occupancy determination, said system comprising:

a processor;

a data bus coupled to said processor; and

a computer-usable medium embodying computer program code, said computer-usable medium being coupled to said data bus, said computer program code comprising instructions executable by said processor and configured for;

acquiring at least one image frame of a parking area of interest from among at least two regions of interest defined with respect to said parking area of interest;

analyzing said at least two regions of interest defined within said parking area of interest for motion detection or content change detection;

performing a vision-based image content classification operation with respect to a first region of interest among said at least two regions of interest based on a result of said content change detection; and

performing an object tracking operation with respect to a second region of interest among said at least two regions of interest if a result of said vision-based image content classification operation indicates a presence of an object of interest within said parking area of interest to thereafter determine a type of access authorized with respect to said parking area of interest and based on a result of said object tracking operation.

9. The system of claim 8 wherein said parking area of interest comprises at least one of a parking lot entrance or a parking lot exit.

10. The system of claim 9 wherein said instructions are further configured for defining said at least two regions of interest with respect to said parking area of interest.

11. The system of claim 10 wherein said instructions are further configured for training said vision-based image content classification with respect to said first region of interest for use in performing said image content classification operation.

12. The system of claim 11 wherein said image content classification includes classification with respect to a presence of at least one vehicle class and without a presence of any vehicle class.

13. The system of claim 11 wherein said instructions are further configured for training a classifier for performing said image content classification by:

analyzing at least two frames of training videos at said first region of interest for motion detection;

automatically collect training samples, an image of said first region of interest in one of said at least two frames, if said motion detection exceeds a threshold; and

grouping collected samples by human operation into two classes comprising said presence of said at least vehicle class and said without said presence of any vehicle class.

14. The system of claim 11 wherein said instructions are further configured for:

smoothing a resulting tracked trajectory with respect to said object tracking operation to generate a smoothed trajectory; and

determining said type of access based on a direction and a travel distance of said smoothed trajectory.

15. A non-transitory processor-readable medium storing computer code representing instructions to cause a hybrid process of video and vision based access control for parking occupancy determination, said computer code further comprising code to:

acquire at least one image frame of a parking area of interest from among at least two regions of interest defined with respect to said parking area of interest;

analyze said at least two regions of interest defined within said parking area of interest for motion detection or content change detection;

perform a vision-based image content classification operation with respect to a first region of interest among said at least two regions of interest based on a result of said content change detection; and

perform an object tracking operation with respect to a second region of interest among said at least two regions of interest if a result of said vision-based image content classification operation indicates a presence of an object of interest within said parking area of interest to thereafter determine a type of access authorized with respect to said parking area of interest and based on a result of said object tracking operation.

16. The processor-readable medium of claim 15 wherein said parking area of interest comprises at least one of a parking lot entrance or a parking lot exit.

17. The processor-readable medium of claim 15 wherein said code further comprises code to define said at least two regions of interest with respect to said parking area of interest.

18. The processor-readable medium of claim 15 wherein said code further comprises code to train said vision-based image content classification with respect to said first region of interest for use in performing said image content classification operation.

19. The processor-readable medium of claim 18 wherein said vision-based image content classification includes classification with respect to a presence of at least one vehicle class and without a presence of any vehicle class.

20. The processor-readable medium of claim 19 wherein said code further comprises code to train a classifier for performing said image content classification by:

analyzing at least two frames of training videos at said first region of interest for motion detection;

automatically collect training samples, an image of said first region of interest in one of said at least two frames, if said motion detection exceeds a threshold; and

grouping collected samples by human operation into two classes comprising said presence of said at least vehicle class and said without said presence of any vehicle class.