Underwater Image Registrtion With Improved SIFT Algorithm

Underwater image registration is widely needed for various applications these days. Research has been carried out widely in the area of image registration. Underwater images are degraded due to various factors. For the feature detection and matching, Scale Invariant Feature Transform (SIFT) is the technique used most of the times. This paper proposes improvements in SIFT in order to achieve better registration of underwater images. Gabor filter is used as prefilter and Hausdorff distance is used to calculate distance between two keypoints. More number of keypoints is detected and proper matching is achieved with this modified algorithm.