Research And Application Of Intelligent Vehicle Tracking Method Based On Motion Background Compensation

The target detection and target tracking technology in the dynamic background is widely used in various aspects,such as drone aerial photography,unmanned driving,precision weapon guidance and so on.Although there is strong practicality in target detection and target tracking in a dynamic context,it is very difficult to implement and is therefore a hot topic of research.Although the traditional detection algorithm is relatively fast,it is not suitable for the detection and tracking of the target in the motion background.There is a certain degree of difficulty in correctly detecting the target.In order to realize the detection and tracking of the target in the sports background,this paper proposes a background-based motion compensation target detection and target tracking algorithm,and applies this method to intelligent vehicle tracking.The method first performs Harris corner detection on two continuous frames of images,then estimates the camera model by matching and filtering out the feature point pairs,and then compensates the background according to the obtained camera model to obtain relatively static background information.Use the frame difference method to detect moving targets.The Camshift algorithm is used to continuously track the moving target.The average value of the H component in the initial search box is used as a reference,and the threshold t is set.When the difference between the mean value of the H component in the search box and the reference is greater than t,the target color is recalculated.A histogram,which uses the average value of the H component in the current search box as a reference,updates the color template in real time.The size of the search box of each frame of the obtained image is compared with the initialization.If the difference is larger than the preset threshold,it is considered that the background of the same color is disturbed.In this case,the search box can be repositioned according to the frame subtraction method.In this paper,the above-mentioned target detection and target tracking are applied to thesmart car,and corresponding experiments are carried out.At the same time,a comprehensive analysis of the method is performed in conjunction with the experimental results.