A realization of FIR system characterizing outline of fingers and its application to person identification

This paper discusses on a realization of FIR system characterizing outline of fingers of a single hand and its application to person identification. First, a hand image is obtained from the image acquisition system. Secondly the normalized outline of the fingers with rotation invariance is extracted from the hand image using KL expansion. Thirdly the horizontal and vertical components of the outline of the fingers are expanded into the Fourier series, respectively. Furthermore FIR system is realized by considering the absolute values of the Fourier coefficients (AFC) as the input and output, respectively. Then a person can be identified by evaluating the difference between the AFC for horizontal components of the outline of the fingers for reference person and the output of the FIR system to the input of AFC for vertical components of the outline of the fingers for the person to be identified. Finally, person identification experiments were performed for 2180 hand images obtained from 109 persons. As the results FRR (false rejection) and FAR (false acceptance) error rates were 1.70% and 0.74%, respectively.