Research On Digital Image Detection Technology Of Bridge Traffic Distribution

With the rapid development of bridge construction and transportation industry, vehicledistribution detection work as an important component. Bridge security detection has receivedmore and more attention. Now the researches on bridge traffic load distribution are still at theprimary stage, so how to quickly,accurately and objectively implement bridge trafficdistribution,it’s very important significance for bridge safety monitoring and assessment.In this paper, by using digital image processing technology, real-time monitor the drivingposition of traffic flow on the deck, and shows the vehicle type, vehicle distance with weight,intelligent analysis of different types of vehicles in the actual distribution of bridge deck. Laya foundation of bridge bearing capacity and dynamic. Researching on image processing ofbridge traffic distribution based on monitoring technology. This paper mainly expounds asfollows:(1) According to the actual needs of the load distribution detection system design, Thispaper introduces the main hardware equipment selection method of CCD camera, studies thetopic related image processing algorithms, draws the image algorithm flow chart processingof this system. Through analysing comprehensive comparison, this paper gives this method inimage processing.(2) The most important vehicle recognition and vehicle distance measurement technologyare introduced in detail. By using the method of Harris corner feature matching based, realizesthe automatic matching model, and proves its feasibility through experiments. Using fuzzy setto extract the lane line based on the theory of Hough transform, realizes the accurate positionof the target vehicle.(3) A thorough study is done on the camera system calibration technology, by studying therelationship between linear geometric model directly the corresponding relationship betweenpixel image coordinates and the real traffic scenes derived coordinates.(4) Introduce the system software development environment, and the design anddevelopment of software, the experimental data are analyzed and thus proves the feasibility ofthe algorithm.