A neural-vision based approach to measure traffic queue parameters in real-time

The real-time measurement of queue parameters is required in many traffic s ituations such as accident and congestion monitoring and adjusting the timi ngs of the traffic lights. Previous methods proposed by researchers for que ue detection are based on traditional image processing algorithms. The meth od proposed here is based on applying the combination of edge detection and neural network algorithms. The edge detection technique is used to detect vehicles and estimate the motion, while neural network is used to measure t he queue parameters. The neural network is trained for various road traffic conditions and is able to provide better results than the traditional imag e processing algorithms. (C) 1999 Elsevier Science B.V. All rights reserved .