MULTISCALE APPROACHES TO MOVING TARGET DETECTION IN IMAGE SEQUENCES

We discuss a novel multiscale approach to the detection of moving obje cts in a sequence of images. The approach is based in part on a multif rame generalization of an optical flow estimation algorithm previously developed by two of the authors. This algorithm provides an extremely efficient multiscale method for estimating optical flow in an image s equence and allows for the temporal accumulation of motion information . Moving objects in a sequence correspond to discontinuities in the tr ue optical flow and, as a result, the residual image associated with t he estimated optical flow can be used as a basis for detecting these o bjects. We propose an approach to detection based on morphological pro cessing of the residual image, and illustrate its potential on real da ta.