The Community for Technology Leaders
Green Image
<p><it>Abstract</it>—Optical flow estimation is discussed based on a model for time-varying images more general than that implied in the work of Horn and Schunk [<ref rid="BIBP084321" type="bib">21</ref>]. The emphasis is on applications where low contrast imagery, non-rigid or evolving object patterns movement, as well as large interframe displacements are encountered. Template matching is identified as having advantages over point correspondence and the gradient-based approach in dealing with such applications. The two fundamental uncertainties in feature matching procedures, whether it is template matching or feature point correspondences, are discussed. Correlation template matching procedures are established based on likelihood measurement. A new method for determining optical flow is developed by combining template matching and relaxation labeling. In this method, a number of candidate displacements for each template and their respective likelihood measures are first determined. Then, relaxation labeling is employed to iteratively update each candidate’s likelihood by requiring smoothness within a motion field. Real cloud images taken from meteorological satellites are used to test the usefulness of this method. It is shown in this application that the new method can deal effectively with the uncertainty of multiple peak (<it>multi-modal</it>) correlation surfaces encountered in template matching. The results show significant improvement when compared to that of the maximum cross correlation (MCC), which has been operationally used for cloud tracking, and to that of the method of Barnard and Thompson, which estimates displacements based on combining point correspondences with relaxation labeling.</p>
Optical flow, dynamic scene analysis, motion estimation, cloud motion fields, template matching, cross-correlation, relaxation labeling.

Q. X. Wu, "A Correlation-Relaxation-Labeling Framework for Computing Optical Flow - Template Matching from a New Perspective," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 17, no. , pp. 843-853, 1995.
91 ms
(Ver 3.3 (11022016))