Finger-knuckle-print for identity verification based on difference images

In this paper, we propose to extract global directional features of finger-knuckle-print based on difference image for identity verification. In order to simplify the formulation for computational complexity reduction, the proposed horizontal and vertical difference images are generated based on matrix projection operation. Subsequently, a Heaviside step function is adopted for image ternarization. Next, we extract Fourier features from these ternary images by means of two-dimensional discrete Fourier transform. Finally, matching between extracted features is performed based on an Euclidean distance comparison. Our experiments on IIT Delhi Finger-Knuckle-Image Version 1.0 database show encouraging results in terms of verification accuracy and computing efficiency.