User Authentication Using Hand Images

Hand geometry is recently realized as the biometric modality for personal authentication. Hand geometry has been adjudged to be one of the safest biometric modalities due to its strong resilience against the impostor attack and also one of the easiest modality as per the acquisition. This paper presents an approach for the biometric based authentication using hand geometry from hand images. The proposed biometric system includes enrollment where the hand images are collected, normalization of image, extraction of normalized contour and extraction of palm region .The ROI extracted after the normalization of image. The contour of the hand region from ROI is used to extract structural features that describe the shape of the hand. These features of the known and the unknown users are matched using support vector machine classifier at the verification stage.