Huzhou Traffic Management System Design And Implementation

With the rapid development of China's national economy, the rapid growth of traffic demand, traffic problems have become more serious. The main issues reflected in the road traffic congestion, chaotic traffic, traffic accidents increase, and serious environmental pollution caused by traffic. At present, the traffic problem has an impact on the socio-economic development and improved living standards of a factor, the traffic problem has been more and more attention.Advanced traffic management system to timely and accurate identification of current and future traffic conditions, to implement reasonable and effective traffic control and induced measures, timely processing of cases, to reduce casualties and property losses, and to induce the driver behavior to avoid the congested area of traffic flow to achieve a smooth traffic flow, reduce regional traffic congestion pressure, saving energy and reducing pollution.In this paper, to build Huzhou traffic information management system as background, a comprehensive analysis of advanced traffic management system, information needs, and traffic information based on the characteristics of the system analysis, traffic flow through the prediction, detection, analysis , dig on traffic flow information, timely and accurate identification of the future and real-time traffic flow conditions, was designed and implemented an integrated, unified traffic information processing platform.System design to event management as the basis, the state recognized as a framework for traffic flow, traffic flow forecasting algorithm is analyzed. Through the use of data mining technology in the non-parametric regression splint fitting, wavelet multi-resolution analysis method, proposed based on multi-resolution analysis of the traffic incident detection algorithm. Using the measured data and simulation data traffic simulation system established separately to validate the algorithm. Experimental results show that traffic flow established state identification algorithms have good performance.