A spectral domain feature extraction scheme for palm-print recognition

In this paper, a spectral feature extraction algorithm is proposed for palm-print recognition, which can efficiently capture the detail spatial variations in a palm-print image. The entire image is segmented into several narrow-width bands and the task of feature extraction is carried out in each band using two dimensional Fourier transform. It is shown that the proposed dominant spectral feature selection algorithm is capable of capturing the variation within the palm-print image, which provides not only the advantage of very low feature dimension but also a very high within-class compactness and between-class separability. Extensive experimentations have been carried out upon different publicly available standard palm-print image databases and the recognition performance obtained by the proposed method is compared with those of some of the recent methods. It is found that the proposed method offers a very high degree of recognition accuracy along with huge computational savings.