The Research On Detection And Tracking Of The Moving Object Based On Image Sequences

Visual information, especially dynamic visual information, takes the most part of environment information that is perceived by human beings. Therefore, perceiving the dynamic visual information in the environment has become an important research field in computer vision. Moving objects detection and tracking is one of the most important issues in applied vision and moving image coding and has wide applications in many fields.The research interest of moving objects detection and tracking is video sequence, i.e. image sequence. Object detection in videos involves verifying the presence of an object in image sequences and possibly locating it precisely. Object tracking is to monitor an object's spatial and temporal changes during a video sequence, including its position, size, shape, etc. These two processes are closely related because tracking usually starts with detecting objects. Due to changes in illumination of the scene, background perturbations, precise detection and tracking of moving objects are still a challenge field of research.Based on the review and analysis of current methods of detecting and tracking moving objects, we lay an emphasis on the research of moving objects detection and tracking in image sequence with static background.Concerning the object detection with static background, current theories of object detection and extraction algorithms are summarized and classified into three classes. Temporal Differential method and Adaptive Background Subtraction method are compared with each other through experimental results. At the same time, Adaptive Background Estimation method is mainly analyzed. What's more, during the implementation of Adaptive Background Subtraction method, the paper proposes an improved scheme on updating background for specific application. The experimental results show that the improved scheme can satisfy the need of system to vast extent and improve the accuracy of detected result.By comparing the object tracking algorithms, we propose a tracking algorithm based on Kalman Filter by setting up mathematical model, which has two key techniques—prediction and update. Prediction's main idea is to take full advantage of correlation between frames, and to predict the possible positions of moving target based on previous data. The matching is just completed in the possible region, which improves its speed due to reduction of waiting region. Update's main idea is that the matched region in current frame is regarded as the template of moving target for next frame and repeats like this, which makes the template updated constantly and has better adaptability to the change of target.The algorithm was emulated in the MATLAB, the experimental result proves the validity and feasibility, and it has good application prospects.