A new object motion estimation technique for video images, based on a genetic algorithm

In the search for lower bit rate image compression and representation, a new video motion estimation technique (VMET), that considers video object translation, as well as rotation, and planar multilayering, is described. This new concept uses a modified multipopulation coevolutionary genetic algorithm (MMCGA), that receives the video objects of segmented reference images, and outputs the corresponding motion and layer information, using object and layer genotypes. Genetic operation strategies of reproduction, crossover, mutation, and dominance are applied recurrently in order to create successive generations of genomes with much better fitness, until convergence, or the maximum allowed number of generations is reached. For the increase of prediction accuracy and convergence speed, a lifetime fitness strategy is used. Simulations with synthetic images have shown very encouraging results with the proposed video motion estimation technique, which competes favorably with respect to the conventional algorithms in accuracy, effectiveness, robustness, simplicity and speed.