Nonlinear correlation for motion estimation in sequences of Markov modeled images

A nonlinear correlation algorithm has recently been proposed for estimating the motion of objects from an image pair. This algorithm requires no a priori information on the number, size, or shape of the moving objects, and does not require feature extraction or segmentation of either image. The algorithm yields information on the number of moving objects, the motion of the objects, the size of the objects, and the centroid of the objects. This paper presents several modifications to this nonlinear correlation algorithm resulting from using a Markov image model. The fundamental equations for implementation and performance analysis are modified to accommodate the Markov model.