On-street vehicle parking occupancy estimation via curb detection
View Patent ↗A method and system for on-street vehicle parking occupancy estimation via curb detection comprises training a computer system to identify a curb, evaluating image data of the region of interest to determine a region wherein a curb is visible in said region of interest, and estimating a parking occupancy of said region of interest according to said region where said curb is visible.
1. A method comprising:
training a computer system to identify a curb;
evaluating image data of a region of interest to determine a region wherein a curb is visible in said region of interest, said evaluation further comprising:
determining a length of said region where said curb is visible;
dividing said length by a predefined length to determine a number of available parking spaces; and
estimating a parking occupancy of said region of interest according to said region where said curb is visible.
2. The method of claim 1 wherein training said computer system to identify a curb further comprises:
collecting data from said scene of interest;
clustering said collected data into groups representative of a selected condition; and
training at least one classifier for each of said groups.
3. The method of claim 2 wherein evaluating image data of said region of interest to determine a region wherein said curb is visible further comprises:
selecting a plurality of classifiers from said at least one classifier for each of said groups;
applying said plurality of classifiers to said region of interest to identify regions wherein said curb is visible; and
combining results from said plurality of classifiers.
4. The method of claim 3 wherein said image data comprises at least one of:
a still photo;
a frame of video data; and
a plurality of frames of video data.
5. The method of claim 3 wherein combining results from said plurality of classifiers comprises at least one of:
averaging all of said results from said plurality of classifiers and comparing said average to a predefined threshold;
comparing each of said results from said plurality of classifiers to a predefined threshold for each of said classifiers; and
performing majority voting on said results from said plurality of classifiers.
6. The method of claim 1 wherein said steps of evaluating image data of a region of interest to determine a region wherein a curb is visible in said region of interest and estimating a parking occupancy of said region of interest according to said region where said curb is visible is done in real-time.
7. A method comprising:
collecting data from a scene of interest;
clustering said collected data into groups representative of a selected condition;
training at least one classifier for each of said groups to identify a curb;
evaluating image data of a region of interest to determine a region wherein a curb is visible in said region of interest, said evaluation further comprising:
determining a length of said region where said curb is visible;
dividing said length by a predefined length to determine a number of available parking spaces; and
estimating a parking occupancy of said region of interest according to said region where said curb is visible.
8. The method of claim 7 wherein evaluating image data of said region of interest to determine a region wherein said curb is visible further comprises:
selecting a plurality of classifiers from said at least one classifier for each of said groups;
applying said plurality of classifiers to said region of interest to identify regions wherein said curb is visible; and
combining results from said plurality of classifiers.
9. The method of claim 8 wherein said image data comprises at least one of:
a still photo;
a frame of video data; and
a plurality of frames of video data.
10. The method of claim 8 wherein combining results from said plurality of classifiers comprises at least one of:
averaging all of said results from said plurality of classifiers and comparing said average to a predefined threshold;
comparing each of said results from said plurality of classifiers to a predefined threshold for each of said plurality of classifiers; and
performing majority voting on said results from said plurality of classifiers.
11. The method of claim 7 wherein said steps of evaluating image data of a region of interest to determine a region wherein a curb is visible in said region of interest, and estimating a parking occupancy of said region of interest according to said region where said curb is visible is done in real-time.
12. A system comprising:
a camera configured to collect image data;
a processor; and
a computer-usable medium embodying computer code, said computer-usable medium being coupled to said processor, said computer code comprising non-transitory instruction media executable by said processor configured for:
training a computer system to identify a curb;
evaluating image data of a region of interest to determine a region wherein a curb is visible in said region of interest, said evaluation further comprising:
determining a length of said region where said curb is visible;
dividing said length by a predefined length to determine a number of available parking spaces; and
estimating a parking occupancy of said region of interest according to said region where said curb is visible.
13. The system of claim 12 wherein said computer code comprising non-transitory instruction media executable by said processor configured for training said computer system to identify a curb further comprises:
collecting data from said scene of interest;
clustering said collected data into groups representative of a selected condition; and
training at least one classifier for each of said groups.
14. The system of claim 13 wherein said computer code comprising non-transitory instruction media executable by said processor configured for evaluating image data of said region of interest to determine a region wherein said curb is visible further comprises:
selecting a plurality of classifiers from said at least one classifier for each of said groups;
applying said plurality of classifiers to said region of interest to identify regions wherein said curb is visible; and
combining results from said plurality of classifiers.
15. The system of claim 14 wherein said image data comprises at least one of:
a still photo;
a frame of video data; and
a plurality of frames of video data.
16. The system of claim 14 wherein said computer code comprising non-transitory instruction media executable by said processor configured for combining results from said plurality of classifiers comprises at least one of:
averaging all of said results from said plurality of classifiers and comparing said average to a predefined threshold;
comparing each of said results from said plurality of classifiers to a predefined threshold for each of said classifiers; and
performing majority voting on said results from said plurality of classifiers.
17. The system of claim 12 wherein said computer code comprising non-transitory instruction media executable by said processor configured for said steps of evaluating image data of a region of interest to determine a region wherein a curb is visible in said region of interest and estimating a parking occupancy of said region of interest according to said region where said curb is visible is done in real-time.