Research On The Technology Of Moving Object Detection And Tracking In Video

Moving object detection and tracking is one of the hot topics in the field of computer vision.It is the key technology of video data processing,which is widely applied in the fields of intelligent transportation,video surveillance,medicine,industry,military and so on.In recent years,scholars at home and abroad have put forward many effective and practical algorithms through a lot of in-depth research.However,due to the complexity of the actual environment,the existing algorithms could not meet the needs of practical applications.Aiming at the existing problems of the algorithms,this paper studies the technology of moving object detection and tracking,and proposes effective and robust algorithms to make up for the shortcomings of the existing algorithms.Firstly,this paper introduces the research status of moving object detection and tracking at home and abroad,and analyzes the technical difficulties in moving object detection and tracking,and discusses the preprocessing techniques used in image processing,and summarizes the commonly used algorithms in moving object detection and tracking.Secondly,this paper focuses on the theory of Gaussian Mixture Model(GMM)and three-frame difference method,then analyzes the advantages and disadvantages of the two algorithms.On this basis,an improved detection algorithm based on three-frame difference method and GMM is put forward.The algorithm uses a way of maximizing the area of the moving object to get a relatively complete objective area,and then combines the foreground image extracted by GMM algorithm to optimize and obtain the final foreground image.The experimental results show that the algorithm effectively overcomes the shadow caused by illumination and the holes problem caused by inter-frame difference method,and reduces the noise interference.Thirdly,by studying the theory of Camshift algorithm,ORB feature point matching and Kalman filter,this paper proposes an improved Camshift algorithm based on ORB feature point matching and Kalman filter.When the object is lost,the algorithm introduces the ORB feature point matching,then obtains the optimal matching point pairs and calculates the number of them.If the number is greater than or equal to the given threshold,the objective position in the current frame is reacquired according to the corresponding relationship between the optimal matching point pairs.Otherwise,the predicted result of Kalman filter which is used as the objective position in the current frame is to ensure the stable tracking.In different experimental environments,the simulation experiments are carried out with the proposed algorithm and the experimental results are analyzed.Finally,this paper designs a moving object detection and tracking system which is built by using MFC class library.It calls the OpenCV functions to apply the proposed detection algorithm and tracking algorithm,and realizes the detection and tracking.