Research On Key Technologies Of Urban Traffic Safety Forewarning System Based On GIS

Traffic safety has become to be the bottlenecks of sustainable development of urban road traffic system, the establishment of GIS-based urban traffic safety forewarning system (GUTSFS) is an effective measure to reduce a series of traffic problems.The GUTSFS integrated the powerful spatial data handling capabilities, rapid positioning,search space, complex inquiry function, effective data integration and visualization of expression means of GIS with the advanced concept in traffic management. First of all, on the basis of in-depth analyzed the causes in traffic accidents, this paper take the analysis, evaluation, detection and prediction based on GIS. Second, we researched on the composition and the module functions of this forewarning system, including data collection and acceptance subsystem, analysis and processing subsystems, query subsystem, accident evaluation, detection, and the forecast subsystem, Alarm and scheduling subsystem. At last, established an incident detection model based on neural network and wavelet. This model fully used the relationship between wavelet transform coefficient and parameter change of traffic stream, and the pattern recognition function of neural network, greatly improved the accuracy of incident detection. This forewarning system can establish space and attribute databases on the basis of different traffic situation in different cities; thereby the intellectualized management level of urban traffic will be improved.