Abstract: The recognition of iris has the potential for high level of accuracy due to its uniqueness and stable feature. In this paper a comparison of SIFT (Scale Invariant Feature Transform) and SURF (Speeded Up Robust Features) algorithm is used to analyse the speed and accuracy of the Iris Recognition system. SIFT and SURF method will extract the local feature points of the iris and then determine the scale invariant key points in the iris image and then express these key points using the local patterns around the key points. The K-nearest neighbour algorithm is used for classification. The experimental result proved that SURF algorithm works faster, efficient and reliable than SIFT.

Keywords: SIFT, SURF, Feature Extraction, K-nearest neighbour.