VIDEO/VISION BASED ACCESS CONTROL METHOD AND SYSTEM FOR PARKING OCCUPANCY DETERMINATION, WHICH IS ROBUST AGAINST CAMERA SHAKE

• US 20150077559A1
• Filed: 09/19/2013
• Published: 03/19/2015
• Est. Priority Date: 09/19/2013
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Wu, Wencheng
• Granted Patent: US 9,736,374 B2
• Time in Patent Office: Days
• US Class Current: 348/148
• 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 20/586 of parking space
• CPC Class Codes: H04N 23/683 performed by a processor, e.g. controlling the readout of an image memory

Abstract

Methods and systems for parking lot occupancy determination. Video of an entrance to a parking lot (or other facility) can be acquired utilizing one or more video cameras focusing on a pre-defined vision-based region of interest and a pre-defined video-based region of interest. The video can be video processed via a classifier that is pre-trained to take into account potential camera shake disturbances to generate video output data thereof. The video output data can then be trajectory smoothed. The video output data (following trajectory smoothing) can then be analyzed for parking lot occupancy data such that the impact of camera shake with respect to the video camera(s) is minimized to determine parking lot occupancy.

Claims

1. A method for parking occupancy determination, said method comprising:

acquiring video of an entrance to a parking lot utilizing at least one video camera focusing on a pre-defined vision-based region of interest and a pre-defined video-based region of interest;

video processing said video via at least one classifier pre-trained to take into account potential camera shake disturbances to generate video output data thereof; and

analyzing said video output data for parking lot occupancy data, wherein an impact of camera shake with respect to said at least one video camera is minimized to determine parking lot occupancy.

2. The method of claim 1 further comprising:

applying distortions that simulate camera shake effects with respect to at least one image frame; and

cropping said at least one image frame at an original region of interest.

3. The method of claim 1 further comprising trajectory smoothing said video output data.

4. The method of claim 1 further comprising tracking at least one vehicle within said video of said entrance to said parking lot utilizing said at least one classifier.

5. The method of claim 3 wherein video processing said video via said at least one classifier further comprises:

video processing said video at said pre-defined video-based region of interest for motion detection and/or tracking, if motion is detected within said pre-defined vision-based region of interest.

6. The method of claim 3 further comprising analyzing said video for a vehicle trajectory of said at least one vehicle to determine if said at least one vehicle is entering or exiting said parking lot.

7. The method of claim 1 further comprising pre-training said at least one classifier based on pulled samples indicative of camera shake in scope to include said potential camera shake disturbances into said at least one classifier.

8. The method of claim 7 further comprising:

randomly moving, rotating, scaling, and/or warping said pre-defined vision-based region of interest; and

cropping said training samples to capture samples indicative of said camera shake.

9. The method of claim 8 further comprising manipulating distortion parameters to generate a plurality of varying levels of simulated camera shake samples for use as said training samples.

10. The method of claim 9 further comprising:

measuring or estimating an amount of said camera shake at least one of image processing with respect to video frames of said video, the use of a gyroscope and/or a motion detection device; and

selecting an appropriate pre-trained classifier among said at least one classifier to take into account said potential camera shake disturbances to generate said video output data thereof.

11. The method of claim 1 further comprising:

measuring changes in a field of view using template matching between templates derived from at least one original scene and at least one current scene with motion removed; and

employing said changes measured in said field of view to adjust data indicative of at least one region of interest for vision-based processing thereof if said changes measured are above a pre-determined threshold so that an impact of said changes with respect so said field of view to performance is minimized and determine said parking lot occupancy.

12. A system for parking occupancy determination, said system comprising:

a processor; and

a computer-usable medium embodying computer program code, said computer-usable medium capable of communicating with the processor, said computer program code comprising instructions executable by said processor and configured for;

acquiring video of an entrance to a parking lot utilizing at least one video camera focusing on a pre-defined vision-based region of interest and a pre-defined video-based region of interest;

video processing said video via at least one classifier pre-trained to take into account potential camera shake disturbances to generate video output data thereof; and

analyzing said video output data for parking lot occupancy data, wherein an impact of camera shake with respect to said at least one video camera is minimized to determine parking lot occupancy.

13. The system of claim 12 wherein said instructions are further configured for trajectory smoothing said video output data.

14. The system of claim 12 wherein said instructions are further configured for tracking at least one vehicle within said video of said entrance to said parking lot utilizing said at least one classifier.

15. The system of claim 14 wherein said instructions are further configured for video processing said video at said pre-defined video-based region of interest for motion detection and/or tracking, if motion is detected within said pre-defined vision-based region of interest.

16. The system of claim 14 wherein said instructions are further configure for analyzing said video for a vehicle trajectory of said at least one vehicle to determine if said at least one vehicle is entering or exiting said parking lot.

17. The system of claim 12 wherein said instructions are further configured for pre-training said at least one classifier based on pulled samples indicative of camera shake in scope to include said potential camera shake disturbances into said at least one classifier.

18. The system of claim 17 wherein said instructions are further configured for:

randomly moving, rotating, scaling, and/or warping said pre-defined vision-based region of interest; and

cropping said training samples to capture samples indicative of said camera shake.

19. The system of claim 18 wherein said instructions are further configured for manipulating distortion parameters to generate a plurality of varying levels of simulated camera shake samples for use as said training samples.

20. A processor-readable medium storing code representing instructions to cause a process for parking occupancy determination, said code comprising code to:

acquire video of an entrance to a parking lot utilizing at least one video camera focusing on a pre-defined vision-based region of interest and a pre-defined video-based region of interest;

video process said video via at least one classifier pre-trained to take into account potential camera shake disturbances to generate video output data thereof; and

analyze said video output data for parking lot occupancy data, wherein an impact of camera shake with respect to said at least one video camera is minimized to determine parking lot occupancy.