A Detection Of Moving Objects Based On Changes In The Regional Segmentation Method

In recent year, content-based image retrieval (CBIR) has been an increasingly active research, CBIR is very useful in various areas, but is extremely difficult to put into use, because the computer has very limited ability in understanding and analyzing the semantic content of the image. 　There are three types of searches CBIR: based on features, on color, and on knowledge. But under most circumstances, what appeal to people is not the entire image but region or object, especially the moving object. Therefore, in recent years, people bring out the image expression and the research method based on moving object, which lay emphasis on the understanding of the main object and the analysis of the video content. In the paper, It make some helpful research and discussions on this point and put forward a practical segment method on moving objects according to moving region detection.The traditional segment method according to moving region detection gets the moving information by the residual quality of space and time between two frame, and then eliminates the disturbance of noises through statistics way,. while ,this method is only acceptable when the camera is fixed. Otherwise it doesn't work. Object detection and segmentation is a key problem in object-based image and video retrieval. The quantity and the target-similarity of the background objects make the detection and segmentation even harder .A practical target detection and segmentation algorithm is presented here .It includes two processing steps. First, to match the scene, extract the movement information, and then perform the first time detection based on it. Second, to perform the fine detection and segmentation based on the time-space relativity of the movement .The low missing rate of the first time detection and the low false rates of the fine detection guarantee the stability of the algorithm. Experiments prove that this method can achieve a satisfactory result, and over come the traditional one's shortcoming of improper for the camera's movement.