Research On Detection And Tracking Algorithm Of Foreign Object Intrusion On Railway Track

In recent years,railway train operation accidents have occurred frequently,causing serious loss of life and property.Research found that foreign matter intrusion on railway tracks is one of the main causes of train accidents.Aiming at moving targets in video surveillance,the paper studies the detection and tracking algorithm of foreign body intrusion in railway tracks based on the existing detection methods,and realizes the accurate detection and tracking of foreign body targets in the track.The main research contents of the thesis are as follows:First,aiming at the adverse effects of haze weather on image quality,based on the dark channel prior algorithm,a defogging algorithm combining bottom hat transformation and image fusion is proposed.Compared with the traditional dark channel prior algorithm,the peak information of the image.The peak signal to noise ratio(PSNR)and structural similarity(SSIM)are increased by 4.3% and 8.5%,respectively,and the processing time is reduced by 98.5%;then grayscale processing is performed to reduce the complexity of the algorithm;then the image is processed by histogram equalization to enhance the image contrast and facilitate the detection and extraction of track edge.Second,when the traditional Canny operator detects the edge of the track,the detection result includes useless edges such as gravel and sleepers around the track,which increases the difficulty of extracting the track edge.In order to solve this problem,the paper uses the Canny operator with adaptive high and low thresholds to detect the edge of the track,and uses the connected domain analysis to eliminate areas with small connected domains in the binary image,so as to detect the edge of the track while successfully eliminating irrelevant scenes.Edge interference;then the hough transform is used to extract the track line and determine the position information of the track,and then define the dangerous area of the track.Third,in order to solve the problems of ghost,dynamic background interference and incomplete detection results,an improved vibe algorithm is proposed.First of all,in the process of initializing the background model by means of the mean value method,the pixel values that have changed greatly are eliminated,so as to obtain a background image close to the real scene,which effectively suppresses the ghost phenomenon;then,an adaptive threshold radius is used instead of a fixed threshold radius,thereby improving the algorithm’s ability to adapt to dynamic backgrounds;finally,median filtering and area filling are used to filter out small noise points and fill the void areas of the detection results respectively.The experimental results in different scenarios show that the improved algorithm in this paper has better detection performance.Forth,a moving target tracking algorithm combining Vi Be and Camshift is used to track the moving target in the video image,and at the same time,it is judged whether the moving target has orbit intrusion behavior according to the delineated orbital dangerous area,and if it does,the color of the tracking frame is changed.Experimental results show that the algorithm can achieve continuous and accurate tracking of moving targets.