Research Of Video Surveillance Based On Panoramic Camera

Video surveillance widely used in many public places for its intuitive, accurate, timely and rich information content. Construction of both intuitive and intelligent video surveillance system is a major area of security research. In this paper, a full study of detection and tracking moving target is done based on the panoramic camera. With the purpose of enhancing the intuition of front-end cameras and enriching the wisdom of back-end video processing, we designed a complete video surveillance system which using both panoramic camera and speed dome.In respect of moving target detection, we effectively detect the moving target with the method of combining Surendra background updating and frame difference.In respect of moving target tracking, an algorithm based on Kalman filter and matching matrix is designed for multiple targets tracking and the results show that it can work fine. In addition, we achieve to control speed dome to shooting the moving target by way of changing panoramic pixel coordinates to PTZ coordinates.With the model of using panoramic cameras and speed dome plus PTZ controller, we design a video surveillance system which not only could get the overall information of the operating room but also could get the local details. Furthermore, features of staff statistics and in post detection are included in this system.