A New Translation Matching Method Based on Autocorrelated Normalized Cross-Power Spectrum

Translation matching is one of the most fundamental problems in the field of image matching, and the normalized cross-power spectrum (NCPS)-based methods have achieved great success regarding this problem. However, when the images to be matched are seriously corrupted by noise, most current NCPS-based methods cannot obtain satisfactory results. Besides, the 2-D phase extraction of the NCPS, which is required in most NCPS-based methods, may cause an additional error to the final result. In this article, we proposed the concept of autocorrelated NCPS (ANCPS) that is theoretically proved to be able to significantly alleviate the influence of noise and developed a new method based on it. Furthermore, by utilizing the property of equal phase interval of ANCPS, the 2-D phase extraction problem is also naturally avoided in our method. The experiments with simulated and real data demonstrate that the presented method has a better performance in both accuracy and antinoise performance compared with state-of-the-art methods.