Side window detection through use of spatial probability maps

• US 9,928,424 B2
• Filed: 02/22/2016
• Issued: 03/27/2018
• Est. Priority Date: 02/22/2016
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Inventors: Bulan, Orhan, Xu, Beilei, Paul, Peter
• Primary Examiner(s): Bella, Matthew
• Assistant Examiner(s): TORRES, JOSE
• Application Number:
  US15/049,568
• Publication Number:
  US 20170243067A1
• Time in Patent Office: 764 Days
• Field of Search: None
• CPC Class Codes: G06F 18/2413 based on distances to training or reference patterns
• CPC Class Codes: G06T 2207/30248 Vehicle exterior or interior
• CPC Class Codes: G06T 2207/30268 Vehicle interior
• 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 20/59 inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
• CPC Class Codes: G06V 20/593 Recognising seat occupancy
• CPC Class Codes: G06V 2201/08 Detecting or categorising vehicles

Abstract

Methods, systems, and processor-readable media for detecting the side window of a vehicle. A spatial probability map can be calculated, which includes data indicative of likely side window locations of a vehicle in an image. A side window detector can be run with respect to the image of the vehicle to determine detection scores. The detection scores can be weighted based on the spatial probability map. A detected region of interest can be extracted from the image as extracted image patch. An image classification can then be performed with respect to the extracted patch to provide a classification that indicates whether or not a passenger is in the vehicle or no-passenger is in the vehicle.

Claims

1. A method for detecting a side window of a vehicle, said method comprising:

calculating a spatial probability map that includes data indicative of likely side window locations of a vehicle in an image;

running a side window detector with respect to said image of said vehicle to determine a plurality of detection scores;

weighting said plurality of detection scores based on said spatial probability map;

extracting a detected region from said image as extracted patch of said image; and

performing an image classification with respect to said extracted patch to provide a classification that indicates whether or not a passenger is in a rear seat of said vehicle or no-passenger is in said rear seat of said vehicle.

2. The method of claim 1 further comprising initially capturing said image of said vehicle utilizing an HO/HOT image acquisition system.

3. The method of claim 1 wherein said calculating said spatial probability map, further comprises:

calculating said spatial probability map by labeling a passenger side window on a set of training images.

4. The method of claim 1 wherein said calculating said spatial probability map, further comprises:

detecting a driver side window of said vehicle in said image; and

referring passenger side window locations on said vehicle in said image to a detected driver side window of said vehicle in said image.

5. The method of claim 1 wherein said calculating said spatial probability map, further comprises:

detecting a windshield of said vehicle from a front view image of said vehicle derived from said image;

performing a feature matching operation between said detected windshield and said side view image of said vehicle; and

referring likely passenger side window locations of said vehicle in said image based on a location of matched features determined as a result of said feature matching operation.

6. The method of claim 1 further comprising training a classifier in an offline phase to perform said image classification.

7. The method of claim 6 wherein said training said classifier further comprises:

training said classifier utilizing said extracted patch to distinguish passenger side window images from types of image patches.

8. The method of claim 6 wherein said running said window detector, further comprises:

running said side window detector utilizing a DPM (Deformable Parts Model).

9. A system for detecting a side window of a vehicle, said system comprising:

at least one processor; and

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

calculating a spatial probability map that includes data indicative of likely side window locations of a vehicle in an image;

running a side window detector with respect to said image of said vehicle to determine a plurality of detection scores;

weighting said plurality of detection scores based on said spatial probability map;

extracting a detected region from said image as extracted patch of said image; and

performing an image classification with respect to said extracted patch to provide a classification that indicates whether or not a passenger is in a rear seat of said vehicle or no-passenger is in said rear seat of said vehicle.

10. The system of claim 9 wherein said instructions are further configured for initially capturing said image of said vehicle utilizing an HOV/HOT image acquisition system.

11. The system of claim 9 wherein said instructions for calculating said spatial probability map, further comprises instructions for:

calculating said spatial probability map by labeling a passenger side window on a set of training images.

12. The system of claim 9 wherein said instructions for calculating said spatial probability map, further comprises instructions for:

detecting a driver side window of said vehicle in said image; and

referring passenger side window locations on said vehicle in said image to a detected driver side window of said vehicle in said image.

13. The system of claim 9 wherein said instructions for calculating said spatial probability map, further comprises instructions for:

detecting a windshield of said vehicle from a front view image of said vehicle derived from said image;

performing a feature matching operation between said detected windshield and said side view image of said vehicle; and

referring likely passenger side window locations of said vehicle in said image based on a location of matched features determined as a result of said feature matching operation.

14. The system of claim 9 wherein said instructions are further configured for training a classifier in an offline phase to perform said image classification.

15. The system of claim 14 wherein instructions for said training said classifier further comprises instructions configured for:

training said classifier utilizing said extracted patch to distinguish passenger side window images from types of image patches.

16. The system of claim 15 wherein said instructions for running said window detector, further comprises instructions configured for:

running said side window detector utilizing a DPM (Deformable Parts Model).

17. A non-transitory processor-readable medium storing computer code representing instructions to cause a process for detecting a side window of a vehicle, said computer code comprising code to:

calculate a spatial probability map that includes data indicative of likely side window locations of a vehicle in an image;

run a side window detector with respect to said image of said vehicle to determine a plurality of detection scores;

weight said plurality of detection scores based on said spatial probability map;

extract a detected region from said image as extracted patch of said image; and

perform an image classification with respect to said extracted patch to provide a classification that indicates whether or not a passenger is in a rear seat of said vehicle or no-passenger is in said rear seat of said vehicle.

18. The processor-readable medium of claim 17 wherein said code is further configured for initially capturing said image of said vehicle utilizing an HOV/HOT image acquisition system.

19. The processor-readable medium of claim 17 wherein said code is further configured for training a classifier in an offline phase to perform said image classification.

20. The processor-readable medium of claim 19 wherein:

said code for training said classifier in said offline phase to perform said image classification further comprises code for training said classifier utilizing said extracted patch to distinguish passenger side window images from types of image patches; and

said code for running said window detector, further comprises code for running said side window detector utilizing a DPM (Deformable Parts Model).