A Distributed Real Time Knowledge-Based System Using Video Image Processing for Junctions Automatic Incident Detection

by S. Sellam, INRETS-MAIA, Arcueil, France,
A. Boulmakoul, INRETS-MAIA, Arcueil, France,
J. C. Pierrelée, INRETS-MAIA, Arcueil, France,

Document Type: Proceeding Paper

Part of: Applications of Advanced Technologies in Transportation Engineering


The scope of the INVAID project (in the European DRIVE program) is to build an automatic incident detection system which incorporate the image processing capabilities. In the urban aspects of this project it is noticed that the output of the image processing is a binary image, i.e., a black point belongs to a vehicle in the image and a white one to the background. A mock-up of this expert system is currently designed with G2 (from GENSYM Corp.) a real time expert system. This paper reports the main components of the system (including the Image Processing main characteristics) and the main elements of the knowledge base leading to a diagnosis process.

Subject Headings: Imaging techniques | Knowledge-based systems | Automation | Traffic surveys | Traffic signals | Signal processing | Intelligent transportation systems | Traffic surveillance

Services: Buy this book/Buy this article


Return to search