초록 ▼

A traffic sensor system for detecting and tracking vehicles is describede. A video camera is employed to obtain a video image of a section of a roadway. Motion is detected through changes in luminance and edges in frames of the video image. Predetermined sets of pixels (“tiles”) in the

A traffic sensor system for detecting and tracking vehicles is describede. A video camera is employed to obtain a video image of a section of a roadway. Motion is detected through changes in luminance and edges in frames of the video image. Predetermined sets of pixels (“tiles”) in the frames are designated to be in either an “active” state or an “inactive” state. A tile becomes active when the luminance or edge values of the pixels of the tile differ from the respective luminance or edge values of a corresponding tile in a reference frame in accordance with predetermined criteria. The tile becomes inactive when the luminance or edge values of the pixels of the tile do not differ from the corresponding reference frame tile in accordance with the is predetermined criteria. Shape and motion of groups of active tiles (“quanta”) are analyzed with software and a neural network to detect and track vehicles.

대표청구항 ▼

1. Apparatus for detecting at least one vehicle, comprising:at least one camera that provides a first video frame that is representative of a field of view of said camera; anda computation unit operative to: (1) organize the first video frame and a second video frame also representative of the field

1. Apparatus for detecting at least one vehicle, comprising:at least one camera that provides a first video frame that is representative of a field of view of said camera; anda computation unit operative to: (1) organize the first video frame and a second video frame also representative of the field of view of the camera into respective pluralities of tiles, each tile being a set of adjacent pixels having predetermined dimensions smaller than a representation of the vehicle in the first video frame, said second video frame being a reference frame representing a combination of previous frames from said camera and being updated by combining selected new frames with said combination of previous frames, said combination of previous frames being weighted more heavily than each said new frame during said update of said second video frame, (2) compare each of a plurality of tiles of the first video frame with a corresponding tile from the second video frame to detect a difference of predetermined magnitude and type,wherein for each tile, if the difference of predetermined magnitude and type is detected, then the tile is in a first state, and if the difference of said predetermined magnitude and type is not detected, then the tile is in a second state, and (3) determine whether a plurality of proximate tiles are in said first state, such a determination indicating that the vehicle is present in the field of view of said camera at a current position represented by said plurality of proximate tiles. 2. The apparatus of claim 1 wherein said second video frame represents a single selected previous frame from said camera. 3. The apparatus of claim 1 wherein said computation unit employs at least one segmenter algorithm to identify at least one feature from each of said plurality of proximate tiles in said first state. 4. The apparatus of claim 1 wherein said computation unit includes a classification engine that is employed to match patterns defined by said plurality of proximate tiles with known patterns to produce a score that indicates a probability that said plurality of proximate tiles represents a vehicle. 5. The apparatus of claim 4 wherein a roadway is within the field of view of said camera and said computation unit adjusts the mapping of video frames to compensate for the grade of the roadway. 6. The apparatus of claim 4 wherein a roadway is within the field of view of said camera and said computation unit adjusts the mapping of video frames to compensate for the bank of the roadway. 7. The apparatus of claim 4 wherein a velocity of said at least one vehicle is calculated from a change in position of the vehicle from said first video frame to said second video frame. 8. The apparatus of claim 7 wherein an anticipated future motion for said at least one vehicle is calculated from said calculated velocity. 9. The apparatus of claim 1 wherein a first zone is defined in said first video frame, and new vehicles that are identified outside of said first zone are disregarded. 10. The apparatus of claim 1 wherein a second zone is defined in said first video frame, and a number of vehicles passing by the camera is determined from the portion of the image within said second zone. 11. The apparatus of claim 1 wherein said difference of predetermined magnitude and type is an edge detection difference calculated by comparing edges in each of said tiles in said first video frame with edges in each said corresponding tiles in said second video frame. 12. The apparatus of claim 1, wherein said difference of predetermined magnitude and type is a luminance difference, wherein the luminance difference is calculated by comparing the average pixel luminance in each tile in said first video frame with average pixel luminance in the corresponding tile in said second video frame. 13. The apparatus of claim 1 wherein at least one feature in said first and second video frames is employed to align the first video frame with the second video frame. 14. The apparatus of claim 1 wherein said computation unit comprises a video capture card. 15. The apparatus of claim 1 wherein said computation unit comprises a frame grabber and software. 16. The apparatus of claim 1 wherein said computation unit comprises a personal computer. 17. The apparatus of claim 1 wherein said computation unit is operative to detect a plurality of vehicles by determining whether corresponding pluralities of proximate tiles are in said first state indicating that the vehicles are present in the field of view of said camera, each said plurality of proximate tiles representing a current position of the corresponding vehicle in the field of view of said camera.