Automated video based ingress/egress sensor reset for truck stop occupancy detection

• US 9,946,936 B2
• Filed: 04/12/2016
• Issued: 04/17/2018
• Est. Priority Date: 04/12/2016
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Inventors: Mizes, Howard A., Kulkarni, Rakesh S., Paul, Peter
• Primary Examiner(s): CHANG, JON CARLTON
• Application Number:
  US15/096,483
• Publication Number:
  US 20170293807A1
• Time in Patent Office: 735 Days
• CPC Class Codes: G06F 18/2411 based on the proximity to a decision surface, e.g. support vector machines
• CPC Class Codes: G06F 18/2415 based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
• CPC Class Codes: G06F 18/2433 Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
• CPC Class Codes: G06T 2207/30264 Parking
• CPC Class Codes: G06V 10/764 using classification, e.g. of video objects
• CPC Class Codes: G06V 20/41 Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items segmenting video sequences G06V20/49
• 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: G06V 20/586 of parking space
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources

Abstract

A method and system for parking occupancy detection comprises collecting video of the parking facility with a video capturing device, counting an occupancy of a parking facility using at least one sensor to establish a sensor occupancy count, classifying each of a plurality of parking spots as occupied or vacant with a classifier according to the classification threshold in order to establish a video occupancy count, determining a difference between the sensor occupancy count and the video occupancy count, and setting the sensor occupancy count to equal the video occupancy count if the difference between the sensor occupancy count and the video occupancy count exceeds a difference threshold.

Claims

1. A method for parking occupancy detection comprising:

collecting video of a parking facility with a video camera;

counting an occupancy of said parking facility using at least one sensor to establish a sensor occupancy count;

classifying each of a plurality of parking spots as occupied or vacant with a classifier according to a classification threshold in order to establish a video occupancy count;

determining a difference between said sensor occupancy count and said video occupancy count; and

setting said sensor occupancy count according to said difference between said sensor occupancy count and said video occupancy count.

2. The method of claim 1 wherein classifying each of a plurality of parking spots as occupied or vacant further comprises:

identifying a region of interest for each of said plurality of parking spots;

randomly distorting said region of interest for each of said plurality of parking spots;

calculating region of interest features for each of said plurality of parking spots wherein said region of interest features indicate if said parking spot is occupied or vacant; and

classifying each or said plurality of parking spots as occupied or vacant based on at least one value for said region of interest features compared to a classification threshold.

3. The method of claim 1 further comprising:

assigning said classification threshold wherein said classification threshold is dependent on said sensor occupancy count.

4. The method of claim 3 wherein assigning a classification threshold further comprises:

establishing an ROC curve for said parking facility;

determining a current sensor occupancy count; and

selecting a threshold value on said ROC curve associated with said current sensor occupancy count.

5. The method of claim 1 further comprising:

logging said sensor occupancy count on a set time interval;

logging said video occupancy count on said set time interval; and

averaging said difference between said sensor occupancy counts and said video occupancy counts over a set time interval for each or said logged sensor occupancy counts and logged video occupancy counts.

6. The method of claim 1 wherein said classifier comprises a computer vision classifier comprising at least one of:

a support vector machine classifier; and

a random forest classifier.

7. The method of claim 1 wherein said at least one sensor comprises an ingress sensor configured at an entry to said parking facility and an egress sensor configured at an exit to said parking facility.

8. The method of claim 4 wherein establishing an ROC curve for said parking facility further comprises:

generating an independent ROC curve for each of said plurality of parking spots; and

generating an independent ROC curve for each of a plurality of video cameras.

9. A system for parking occupancy detection comprising:

a video camera for collecting video of a parking facility;

a sensor for counting an occupancy of said parking facility used to establish a sensor occupancy count;

a processor; and

a non-transitory computer-usable medium embodying computer code executable by said processor, said computer code configured for;

classifying each of a plurality of parking spots as occupied or vacant with a classifier according to a classification threshold in order to establish a video occupancy count;

determining a difference between said sensor occupancy count and said video occupancy count; and

setting said sensor occupancy count according to said difference between said sensor occupancy count and said video occupancy count.

10. The system of claim 9 wherein classifying each of a plurality of parking spots as occupied or vacant further comprises:

identifying a region of interest for each of said plurality of parking spots;

randomly distorting said region of interest for each of said plurality of parking spots;

calculating region of interest features for each of said plurality of parking spots wherein said region of interest features indicate if said parking spot is occupied or vacant; and

classifying each or said plurality of parking spots as occupied or vacant based on at least one value for said region of interest features and a classification threshold.

11. The system of claim 9 wherein said computer code is further configured for:

assigning said classification threshold wherein said classification threshold is dependent on said sensor occupancy count.

12. The system of claim 11 wherein assigning a classification threshold further comprises:

establishing an ROC curve for said parking facility;

determining a current sensor occupancy count; and

selecting a threshold value on said ROC curve associated with said current sensor occupancy count.

13. The system of claim 9 wherein said computer code is further configured for:

logging said sensor occupancy count on a set time interval;

logging said video occupancy count on said set time interval; and

averaging said difference between said sensor occupancy counts and said video occupancy counts over a set time interval for each or said logged sensor occupancy counts and logged video occupancy counts.

14. The system of claim 9 wherein said classifier comprises a computer vision classifier comprising at least one of:

a support vector machine classifier; and

a random forest classifier.

15. The system of claim 9 wherein said at least one sensor comprises an ingress sensor configured at an entry to said parking facility and an egress sensor configured at an exit to said parking facility.

16. The system of claim 12 wherein establishing an ROC curve for said parking facility further comprises:

generating an independent ROC curve for each of said plurality of parking spots; and

generating an independent ROC curve for each of a plurality of video cameras.

17. A method for parking occupancy detection comprising:

collecting video of a parking facility with a video camera;

counting an occupancy of said parking facility using at least one sensor to establish a sensor occupancy count;

assigning a classification threshold wherein said classification threshold is dependent on said sensor occupancy count;

establishing an ROC curve for said parking facility;

determining a current sensor occupancy count;

selecting a threshold value on said ROC curve associated with said current sensor occupancy count;

classifying each of a plurality of parking spots as occupied or vacant with a classifier according to said classification threshold in order to establish a video occupancy count;

determining a difference between said sensor occupancy count and said video occupancy count; and

setting said sensor occupancy count according to said difference between said sensor occupancy count and said video occupancy count.

18. The method of claim 17 wherein classifying each of a plurality of parking spots as occupied or vacant further comprises:

identifying a region of interest for each of said plurality of parking spots;

randomly distorting said region of interest for each of said plurality of parking spots;

calculating region of interest features for each of said plurality of parking spots wherein said region of interest features indicate if said parking spot is occupied or vacant; and

classifying each or said plurality of parking spots as occupied or vacant based on at least one value for said region of interest features and said classification threshold.

19. The method of claim 17 further comprising:

logging said sensor occupancy count on a set time interval;

logging said video occupancy count on said set time interval; and

averaging said difference between said sensor occupancy counts and said video occupancy counts over a set time interval for each or said logged sensor occupancy counts and logged video occupancy counts.

20. The method of claim 17 wherein said at least one sensor comprises an ingress sensor configured at an entry to said parking facility and an egress sensor configured at an exit to said parking facility.