Abstract
AbstractAn automatic vehicle speed measurement system is presented that can reliably associate the vehicle ID with the retrieved vehicle speed. The system captures images of vehicles passing between two points on the ground that are in the same vicinity, using a pair of linescan cameras installed in a parallel configuration, and measures the speed of passing vehicles by measuring the time taken for the vehicle license plate to move between the two points. In the proposed method, a pair of linescan cameras arranged in parallel is first used to detect passing vehicles and retrieve images of the front sections of the vehicles by capturing images of the vehicles from an angle diagonal to the direction of motion. High‐resolution images of the front section of the vehicles are obtained by image capture at a diagonal to the direction of the vehicles in order to obtain high spatial and temporal resolution images of the objects passing in front of the linescan cameras without the images being overly large. However, the license plate in the retrieved images is distorted depending on the position of the plate and the vehicle speed, and this distortion is corrected by performing a geometrical transformation on the images, based on the position at which the license plate passes the cameras and the time the vehicle takes to pass, as determined by finding corresponding blocks. After this, license plate recognition is used to determine the vehicle speed at the position of the license plate. A 24‐hour vehicle speed measurement experiment that used an infrared LED for illumination only during the nighttime achieved an 85.1% success rate in measuring the speeds of vehicles at the position of the license plates, thus verifying the validity of the proposed method. © 2006 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 89(10): 33–45, 2006; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjc.20274
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have