Vision-Based Traffic Surveillance System on the Internet

• Venue: ICCIMA '99 Proceedings of the 3rd International Conference on Computational Intelligence and Multimedia Applications
• Year: 1999

Many researches are currently in progress on the application of machine vision to traffic surveillance system. We present a vision-based traffic surveillance system on the Internet. Our system consists of FPAs (Field Processing Agents) and TSM (Traffic Surveillance Manager). The FPAs and TSM both communicate with each other on the Internet. The FPA is a Web server, which serves real-time video data, vehicle velocity and vehicle density when the TSM requests them. An FPA consists of a communication module and an image-processing module. The image-processing module in the FPA performs traffic parameter estimation. Traffic parameters are estimated by just processing groups of image pixels without any understanding of the image. Although this has an accuracy limitation, it can be used in applications requiring an approximate vehicle density and vehicle velocity.