Induction Loop Detection of Bicycles Flow Based on Pattern Recognition (PR) Techniques

Abstract Bicycles flow detection and recognition is a very important factor which will affect the efficiency of city traffic control systems in China. However, to develop a reliable bicycles flow detecting device, as known, is still a knotty problem needed to be solved. The authors present a novel detecting system, in which the signals of bicycles flow is picked-up by buried inductive loop and the classification of bicycles flow is determined by using pattern recognition technique. A prototype of this detecting system has been successfully operated, together with a real-time traffic control system developed by Tianjin University, at an intersection in Tianjin for more than 4 years. In this paper, the detecting principle of bicycles flow and the design of sensor subsystem are presented firstly. Then the characteristics of signal transmission structure from sensors to the host instrument is described in detail. Finally, the pattern recognition method of bicycles flow and the experimented results are also given.