Image Matching for Workpiece Based on Genetic Algorithm

Traditional image matching algorithms are always global search method, but they are time-consuming. In order to meet the real-time requirements of recognition system based on matching edge of the workpiece, a new image matching algorithm which combines improved adaptive genetic algorithm based on information of population generation and improved Hausdorff distance algorithm is proposed here. The use of genetic algorithm search mechanism of the non-ergodicity makes it rapidly converge to the global optimal solution and increases the speed to searching for matching. The experimental results show that the algorithm not only speeds up the matching process, improves the anti-noise performance, but also effectively solves the problem of image matching and recognition in the case of translation, rotation and partial occlusion.