自行车群体流的在线实时测量

本文提出了一种利用环形线圈采用动态频差法配合模式识别技术实现对自行车群体流自动检测的方法。讨论了该方法的基本原理、识别模型及据此方法研制成的在线实时智能式自行车群体流量测量仪的试验结果，其测量准确度达８５％以上，数据以５秒时间间隔存入内存或上位机。此样机在现场连续运行３年多，已获满意结果。

Bicycles Flow On-line and Real-time Detection

For urban traffic management and city planing, on-line real-time accurate auto-matic detection of bicycles flow is very important- However, to develop a reliable bicycles flow detecting device, as know, is still a knotty Problem needed to be solved. In this paper, the dy- namic difference frequency method with Pattern Recognition (PR) technique by using induc-tive loop is presented. The basic principle, the recognition models and the testing results of the detecting system for bicycles flow are given. The detecting system based on this method is op-erating on an intersection for more than 3 years and the results are satisfactory.