Research Of Online Calibration Of Camera With Changeable Extrinsic Parameters In Visual Surveillance

Intelligence visual surveillance is currently one of the most active research topics in computer vision. Visual surveillance automatically analyses the video sequences, and attempts to detect, track and recognize certain objects. Based on this, the visual surveillance system can understand and analyze the objects behaviors, in order to do daily management and reflect when unusual conditions occur.In order to meet with the need of visual surveillance, this paper put forward a thought and means of online calibration of camera with changeable extrinsic parameters, based on machine vision in which a number of digit image disposal technology and pattern identification theory are used. The main contents can be summarized as following:1. A special calibration template with structure of mosaic is designed. There are significant differences between this kind of template and background. So, it's easier to detect feature.2. According to the special structure of the template, first obtain the template in image by detecting the connecting area with structure of mosaic, then find corners of the template using Susan corner detection algorithm. This method can remove most of the background points and enhance the reliability of corner detecting algorithm.3. Tsai Calibration algorithm is improved, and the information which the producer of cameras must give is not necessary any more. The method is simply and accurate highly.4. A method of online calibration of camera with changeable extrinsic parameters is proposed, based on the knowledge of quaternion and real time control of rotation of camera using DSP. The result of experiments shows that this method is efficient and has good precision. It can meet with the need of object localization in visual surveillance system.