Complex Background Traffic Statistical Research

With the development of city and economy of our country,the burden of the road transportation system becomes higher and higher.Therefore,more attention has been paid to the vision-based intelligent transportation system.Research on the technology for the vehicle detection based on image processing is just the hot of the field.In order to identify and track vehicles,vehicles segmentation must be accurate from the background. Therefore,to identify the number and characteristics of vehicles and the type of mobile in same video domain,an important real-time basis can be provided by the joint application of vehicle video detection technology and image processing technology.The concept and fundamentality of Intelligent Traffic System,the importance of image processing technology in vehicle identification and statistic and the summary of vehicle video detection methods are introduced in the paper firstly.Then,video detection technology and other detection technology are compared,and superiority of video detection technology is proposed.The method of background subtraction is mostly studied to pick up the vehicle in the paper.And area method is proposed to add up traffics and identify vehicles.Some image processing basic knowledge related with video detection such as correlated gray-scale,filtering,image enhancement and sharpening and so on are used in the process of research algorithm,and algorithms of the knowledge are compared.And then,image segmentation,vehicle detection and extraction algorithm include threshold segmentation,edge detection and morphological operations are used in the paper.At last a real-time and reliable vehicle recognition and statistic research based on amalgamation of background subtraction and area method is designed and realized through programming with MATLAB/GUI in Windows operating system.In the paper, the algorithm was used to study multi-frame images of traffic flow.The experimental results show that the effect of algorithm on the identification and statistics of traffic flow is very good.