Research And Implementation Of Detection And Tracking Moving Object In Image Sequenc Based On OpenCV

The rapid development of the video surveillance system has been promoted by the rapid development of the computer network and the application requirement of security market, which promotes the monitoring system being widely used in the field of traffic, security and security. Moving object detection and tracking technology is the base of video surveillance and video analysis, and its improvement and development are of great significance to the promotion and application of video surveillance.In view of the current situation of the development of the technology of moving target detection and tracking, this thesis chooses the moving objects in video sequences as the target and preprocess by a variety of digital image processing technology, improves the target detection algorithm to improve the real-time and effectiveness of the detection process, and optimizes the tracking algorithm to achieve the fast and accurate tracking of target based on the OpenCV computer vision library and visual studio 2013 development platform. This thesis mainly includes the following three aspects of research content:In image processing stage, this thesis uses histogram equalization to obtain better visual effect and study the noise elimination effect of the Gaussian blur and median fuzzy to common image, performs threshold segmentation by using Ostu threshold processing method, and improve the binary image and the target structure effect using mathematical morphological filter.Considering the real-time and effectiveness of target detection in the stage of target detection, use the improved background subtraction method by combining the frame difference method to execute the simulation experiment of moving target detection. The detection results of target in the visible light and infrared video show that the algorithm can effectively suppress the background’ interference to the target, and achieve the goal of accurate and complete detection.In the process of target tracking, this thesis mainly studies the Camshift algorithm which tracks target using color feature. Aiming at the deficiency of Camshift algorithm processing the serious occlusion’ target, a Kalman filter is proposed to estimate the target position to modify the Camshift algorithm to track the target. The simulation results show that the Camshift tracking algorithm can accurately track the object with serious occlusion after the position prediction and correction of the target by Kalman, and it has good feasibility and real-time performance.The experimental results show that the improved algorithm of moving object detection and tracking technology has good real-time performance and effectiveness, and it has a wide range of application value for smart city and safe city construction.