| According to the national highway traffic is in urgent need of intelligent management, this paper presents a highway over-speed snapping system based on video detection technique. The whole system monitors the outside through the industrial control computer which equipped with relevant software to work together,it forms a complete management system of highway traffic intelligence,which can effectively achieve the capture job of the highway over-speed vehicles.In order to overcome the defects brought by the traditional sense coil and the radar speed measure method, the paper uses video speed measure methods, namely, to use running vehicle detection algorithm, through software programming to measure speed.In the paper, video trigger and speed measure module use virtual coil detection methods, in order to overcome the problems of error detection, missing detection and low speed accuracy caused by the perspective phenomenon. This paper uses a dual-CCD camera to detect the existence of moving vehicles in the detected region of the two virtual coils separately, to guarantee the stability and reliability of the over-speed snapping system.In the detection algorithm of over-speed snapping system, the paper uses the combination method of background difference and inter-frame difference to achieve the existence detection of the objectives, it can improve the phenomenon of inanition and splitting caused by less internal texture in the inter-frame difference process, experimental result shows that this method can quickly and efficiently extract the moving vehicles under complex traffic scenes, the shadow of the vehicle can lead to detected target area contains redundant shadow region, which will influence the test results.In allusion to this issue, this paper uses the combination algorithm of based on edge detection operator and the shadow part to detect and remove the shadow, then to verify the proposed method through experiment and obtain satisfactory result. |
PDF Full Text Request