METHOD FOR DETECTING LARGE SIZE AND PASSENGER VEHICLES FROM FIXED CAMERAS

• US 20140376769A1
• Filed: 06/20/2013
• Published: 12/25/2014
• Est. Priority Date: 06/20/2013
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Bulan, Orhan, Wu, Wencheng, Wang, Yao Rong, Loce, Robert P.
• Granted Patent: US 9,317,752 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69

Abstract

A method for detecting parking occupancy includes receiving video data from a sequence of frames taken from an associated image capture device monitoring a parking area. The method includes determining at least one candidate region in the parking area. The method includes comparing a size of the candidate region to a size threshold. In response to size of the candidate region meeting and exceeding the size threshold, the method includes determining whether the candidate region includes one of at least one object and no objects. The method includes classifying at least one object in the candidate region as belonging to one of at least two vehicle-types. The method further includes providing vehicle occupancy information to a user.

Claims

1. A method for detecting a vehicle, the method comprising:

receiving video data from a sequence of frames taken from an associated image capture device monitoring a parking area;

determining at least one candidate region in the parking area;

comparing a size of the candidate region to a size threshold;

in response to the candidate region meeting and exceeding the size threshold, determining whether the candidate region includes one of at least one object and no object; and

,

classifying an object in the candidate region as belonging to one of at least two vehicle types.

2. The method of claim 1 wherein the classifying includes:

classifying the object as one of a motorcycle, a passenger vehicle and a commercial vehicle.

3. The method of claim 1, wherein the classifying includes:

in response to the size meeting and exceeding the size threshold, performing a sliding-window search in the candidate region and applying at least one extracted object feature into a first type vehicle classifier to classify the object as being one of a first type vehicle and not one of the first type vehicle; and

,

in response to the size not meeting and exceeding the size threshold, performing a sliding-window search in the candidate region and applying the at least one extracted object feature into a second type vehicle classifier to classify the object as being one of a second type vehicle and not one of the first and second type vehicles.

4. The method of claim 3, wherein the classifying includes:

in response to the at least one object feature matching sample features in the first type vehicle classifier, classifying the vehicle as the first type vehicle; and

in response to the at least one extracted feature not matching the sample features in the first type vehicle classifier, applying the at least one extracted object feature to the second type vehicle classifier.

5. The method of claim 4, wherein the applying the at least one extracted object feature into the first type vehicle classifier includes:

receiving a score from the first type vehicle classifier corresponding to a degree that the at least one object feature matches the sample features in the first vehicle type classifier;

comparing the score to a confidence threshold;

in response to the score meeting and exceeding the confidence threshold, determining that the at least one object feature matches the sample features; and

,

in response to the score not meeting the second threshold, determining that the at least one object feature does not match the sample features.

6. The method of claim 3, wherein the classifying includes:

in response to the at least one extracted object feature matching sample features in the second type vehicle classifier, classifying the object as being the second type vehicle; and

,

in response to the at least one extracted object feature not matching the sample features in the second type vehicle classifier, classifying the object as not being one of the first and second type vehicles.

7. The method of claim 1, wherein the determining the candidate region includes:

tracking a vehicle into a parking spot using a coherent cluster of motion vectors whose length go to zero in the parking region of interest; and

,

identifying a collection of pixels that were last associated with the coherent cluster of motion vectors.

8. The method of claim 1, wherein the classifying includes:

extracting at least one feature from the object using a binary mask;

calculating a size of the object using the at least one extracted feature; and

,

applying the at least one feature to one of a first and second vehicle type classifier based on the size.

9. The method of claim 3, wherein the calculating the size includes one of calculating a physical length of the candidate region using real-world coordinates mapped from pixel coordinates and calculating an area of the candidate region on the image plane.

10. The method of claim 1 further comprising providing notification to a user of vehicles located in the region of interest.

11. A computer program product comprising tangible media which encodes instructions for performing the method of claim 1.

12. A monitoring system for detecting a vehicle, the system comprising:

a vehicle detection device including;

video buffering module for receiving video data from a sequence of frames taken from an associated image capture device monitoring a parking area;

a candidate region determination module for determining at least one candidate region in the parking area;

a vehicle detection and classification module for detecting and classifying at least one object in the candidate region as belonging to one of at least two vehicle-types, the vehicle classification module adapted to;

compare a size of the candidate region to a size threshold, and

in response to size of the candidate region meeting and exceeding the size threshold, determine whether the candidate region includes determining whether the candidate region includes one of at least one object and no object; and

,

a memory for storing the modules and a processor for executing the modules.

13. The system of claim 12, wherein the vehicle classification module is adapted to:

classify the object as one of a motorcycle, a passenger vehicle and a commercial vehicle.

14. The system of claim 12, wherein the vehicle determination and classification module is adapted to:

in response to the size meeting and exceeding the size threshold, perform a sliding-window search in the candidate region and apply at least one extracted object feature into a first type vehicle classifier to classify the object as belonging to one of a first type vehicle and not one of the first type vehicle; and

,

in response to the size not meeting and exceeding the size threshold, perform a sliding-window search in the candidate region and apply the at least one object feature into a second type vehicle classifier to classify the object as belonging to one of a second type vehicle and not one of the first and second type vehicles.

15. The system of claim 14, wherein the vehicle determination and classification module is adapted to:

in response to the at least one object feature matching sample features in the first type vehicle classifier, classify the object as the first type vehicle; and

in response to the at least one extracted feature not matching the sample features in the first type vehicle classifier, apply the at least one object feature to the second type vehicle classifier.

16. The system of claim 15, wherein the vehicle determination and classification module is further adapted to:

receive a score from the first type vehicle classifier corresponding to a degree that the at least one object feature matches the sample features in the first vehicle type classifier;

compare the score to a confidence threshold;

in response to the score meeting and exceeding the confidence threshold, determine that the at least one object feature matches the sample features; and

,

in response to the score not meeting the second threshold, determine that the at least one object feature does not match the sample features.

17. The system of claim 14, wherein the vehicle determination and classification module is adapted to:

in response to the at least one object feature matching sample features in the second type vehicle classifier, classify the object as belonging to the second type vehicle; and

,

in response to the at least one object feature not matching the sample features in the second type vehicle classifier, classify the object as not belonging to one of the first and second type vehicles.

18. The system of claim 12, wherein the candidate region determination module is adapted to:

track a vehicle into a parking spot using a coherent cluster of motion vectors whose length go to zero in the parking region of interest; and

,

identify a collection of pixels that were last associated with the coherent cluster of motion vectors.

19. The system of claim 18, wherein the vehicle determination and classification module is adapted to:

extract at least one feature from the object using the binary mask;

calculate a size of the object using the at least one extracted feature; and

,

apply the at least one feature to one of a first and second vehicle type classifier based on the size.

20. The system of claim 14, wherein the vehicle determination and classification module is adapted to calculate the size by one of calculating a physical length of the candidate region using real-world coordinates mapped from pixel coordinates and calculating an area of the candidate region on the image plane.