Robustness of the Scale Invariant Feature Transform (SIFT) Method in Automatic Ship Recognition

In this paper, an automatic ship recognition method from visible images that uses a class of local invariant features, named Scale Invariant Feature Transform (SIFT), issuggested. These features are invariant to image scaling, translation, and rotation, and partially invariant to illumination changes and occlusions.These features are used as input to a nearestneighbour indexing method that identifies candidate ship matches. Experimental resultsshow that robust ship recognition can be achieved in cluttered partially-occluded naval visible images