Moving object segmentation from video sequences: an edge approach

This paper presents an automatic algorithm for segmenting and extracting moving objects suitable for indoor and outdoor video applications, where the background scene can be captured beforehand. Since edge detection is often used to extract accurate boundaries of the image's objects, the first step in our algorithm is accomplished by combining two edge maps that are detected from the frames difference and background subtraction. After removing edge points that belong to the background, the resulting moving edge map is fed to the object extraction step. A fundamental task in this step is to declare the candidates of the moving object, followed by applying morphological operations. The algorithm is implemented on different kinds of video sequences and good segmentation results are achieved.