Adaptive Control Of Single Intersection Signal Control System

With the rapid development of the global economy and the rapid expansion of city size, continuously strengthen the functions of cities,urban residents have become increasingly frequent economic and cultural activities, thereby facilitating the rapid growth of urban transport demand. However, the rapid development of urban transport in the same time, it also brings a series of problems. The aggravation of urban traffic congestion become a social issue for the development of cities over the world, as it causes frequent traffic accidents, increased vehicle delay, and worsening environment. Intelligence Transport System can improve traffic efficiency greatly as it use information technology, control technology and communication technology, traffic congestion can be solved efficient and economical. This thesis focuses on the fundamental theories and the key technologies of vehicle detection and traffic signal control system based on video sequences in Intelligent Transportation System. The main achievements in the thesis include:Firstly, introduce the history of development of intelligent transportation systems and urban traffic control theory. And make a discussion on some difficulty and problem in the field including theory analysis and application. We also introduce the research of computer vision in intelligent transportation system and the development direction on intelligent traffic system in our country.Secondly, in-depth study based on the video frequency examination transportation information acquisition technology and the digital image processing technology. Introduces a prospect based on edge information extraction method. Extracting accurate background edge of current frame by background edge extracting model, and eliminate it, and ascertain the foreground regions to divide up vehicles. The algorithm is simple and effective for complex urban traffic environment, the lens can effectively deal with jitter, the road edge on both sides of the trees swaying, light changes and other adverse factors caused the vehicle noise testing. For the characteristics of traffic scenes, we proposed a new model of vehicle detection. We construct a new loop-based virtual model of vehicle detection, reduce the range of testing. The algorithm greatly reduces the time in order to ensure the system of real-time.Finally, aiming at the problem that it is difficult to achieve the dynamic optimization for traffic signal in intelligent traffic signal control system. Focuses on how to achieve an efficient and reasonable intersection traffic signal control. For the four-phase intersection, describes the composition and working principle of the system, and gives the software and hardware design. Based on the traffic flow data collected by the CCD sensor, using the optimized control strategy for automatic adjustment of signal timing cycle and the proportion of traffic lights, road traffic junctions to minimize stranded vehicles to achieve an intelligent traffic light controlExperimental results show this system can effectively reduce the number of stops of vehicles and delays, reduce the phenomenon of urban road traffic congestion and greatly improve the capacity of urban intersection and realize the optimal control of in traffic signal.