Method and apparatus for traffic incident detection

• US 6,411,328 B1
• Filed: 11/06/1997
• Issued: 06/25/2002
• Est. Priority Date: 12/01/1995
• Status: Expired due to Term

• Current Assignee: Southwest Research Institute
• Original Assignee: Southwest Research Institute
• Inventors: Franke, Ernest A., Magee, Michael J., Seida, Steven B., Kinkler, Ernest S.
• Primary Examiner(s): BRITTON, HOWARD W
• Application Number:
  US08/968,916
• Time in Patent Office: 1,692 Days
• Field of Search: H04/N.718, 348/143, 348/144, 348/148, 348/149, 348/164, 348/169, 382/103, 382/104, 701/207, 701/117, 701/118, 701/119, 340/937
• US Class Current: 348/149
• CPC Class Codes: G08G 1/04 using optical or ultrasonic detectors
• CPC Class Codes: H04N 23/23 from thermal infrared radiation
• CPC Class Codes: H04N 5/33 Transforming infrared radiation cameras or camera modules for generating image signals from infrared radiation H04N23/20; circuitry of SSIS for transforming infrared radiation into image signals H04N25/20
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Abstract

Apparatus and methods effective for detecting, locating and characterizing traffic-related incidents are based upon improved image processing techniques applied to infrared and visible light spectrum roadway images in a time sequence. Substantially real-time isolation and identification of anomalous or unexpected traffic conditions allows control of traffic signals and dispatch of law enforcement, maintenance or emergency medical care resources to reduce cost and increase safety. Detectable traffic-related incidents include, for example, the appearance of a stationary object on a roadway, the appearance of a pedestrian on a roadway, and the identification and location of vehicles obstructing traffic flow by moving too slowly or erratically or in the wrong direction. Programmable digital computer image processing allows automatic classification of the severity of detected incidents based on the potential for further traffic hazards and injuries, and can support automatic signals to alter traffic flow patterns for reducing the likelihood of incidents, as well as automatic alerts to law enforcement, maintenance and emergency medical personnel.

Claims

1. A method for recognizing a vehicle in a roadway, the method comprising:

acquiring an infrared image;

locating within the roadway image a thermal emission pair having an area of relatively low thermal emission adjacent to and above an area of relatively high thermal emission; and

recognizing a vehicle as being indicated by said thermal emission pair.

2. The method of claim 1, further comprising the step of further processing the infrared image proximate to the thermal emission pair to determine a distinctive feature associated with a vehicle.

3. The method of claim 1, wherein the identifying step is performed by detecting long wave infrared radiation.

4. The method of claim 1, further comprising the step of further processing the thermal emission pair to determine the orientation of the vehicle.

5. The method of claim 1, further comprising the steps of acquiring a visual image of the roadway and of comparing the visual image to the infrared image.

6. The method of claim 5, wherein the comparing step is performed such that it determines whether a vehicle has active headlights.

7. The method of claim 5, wherein the comparing step is performed such that it determines the orientation of a vehicle.

8. The method of claim 5, further comprising the step of determining the three dimensional field of view of the infrared image relative to the visual image.

9. The method of claim 1, further comprising the steps of repeating the acquiring, locating, and recognized steps for a series of images, and of processing the images to determine the motion of the vehicle.

10. The method of claim 9, wherein the processing step is performed by removing moving objects from the series of images and comparing the resulting image to a reference image.

11. A method of recognizing the traffic lane portion of a roadway, comprising the steps of:

acquiring a series of infrared images of the roadway;

in each image of the series, locating a vehicle by detecting within the image a thermal emission pair having an area of relatively low thermal emission adjacent to and above an area of relatively high thermal emission;

tracking the motion of the thermal emission pair;

identifying the traffic lane as the path followed by the thermal emission pair.

12. The method of claim 11, wherein the identifying step is performed by detecting long wave infrared radiation.

13. The method of claim 11, further comprising the step of processing the images to remove moving objects from the images, thereby obtaining a filtered image.

14. The method of claim 11, further comprising the step of comparing the processed image to a reference image to determine the presence of a stationary object in the filtered image.

15. The method of claim 11, further comprising the step of detecting an object outside the vehicle path.

16. The method of claim 11, further comprising the step of repeating the locating and tracking steps to identify a thermal emission pair moving erratically within the traffic lane.

17. The method of claim 11, further comprising the step of repeating the locating and tracking steps to identify a thermal emission pair and the step of processing the thermal emission pair to determine its orientation within the traffic lane.

18. The method of claim 11, further comprising the step of repeating the locating and tracking steps to identify a thermal emission pair moving the wrong direction within the traffic lane.

19. The method of claim 11, further comprising the step of repeating the locating and tracking steps to estimate the speed of a subject thermal emission pair.

20. The method of claim 19, wherein the estimating step is performed by determining the path length and travel time of the subject thermal emission pair.

21. The method of claim 11, wherein the tracking step is performed by labeling at least one pixel representing the vehicle, forming a vehicle composite image comprised of the series of images and containing the labeling pixels, and identifying a curvilinear form in the composite image.

22. A traffic incident detection system, comprising:

an infrared detector; and

a processing system having a processor and processor memory, operable to receive image data signals of a roadway from the infrared detector, and further operable to process the image data signal by locating a thermal emission pair having an area of relatively low thermal emission adjacent to and above an area of relatively high thermal emission, and to identify the thermal emission pair as a vehicle.