A general transformation equation is derived and applied to dynamic and static image differences. It uses translation, rotation, scale, and velocity changes as fundamental object features for image segmentation. A linear minimization algorithm is derived based on a spatial smoothness constraint. The transformation and algorithm constitute an automatic target recognition preprocessor architecture.
John L. Johnson, "Toward a fundamental image preprocessor," Appl. Opt. 29, 4819-4825 (1990)