Cuckoo Search-based range image registration for free-form surface inspection

3D parts inspection can be conducted by comparing the ideal design geometry with the real measurement points. Since the design coordinate system is different from the measurement coordinate system, these measurement points should be registered to the design coordinate system first. In this research area, iterative closest point (ICP) is the best-known algorithm, however, in order to converge to the global minima, ICP needs the good initial parameter which is hard to get in the actual inspection process. In this research, a hybrid Cuckoo Search (CS) method is proposed to solve the registration problem and two different optimizing strategies based on CS are described. The proposed algorithm seems much superior to other algorithms in terms of accuracy and robustness. Experiment results show that the proposed algorithm is effective.