The development of a SOPC for matching fingerprint images is presented in this paper. The major motivation for this work is to develop a fingerprint image matching handheld module, which can overcome the limitation of original phase correlation based algorithm and has more reliability performance to rotated and noisy images. By aid of Fourier-Mellin transform (FMT), frequency domain features are extracted and compared by peak value and distribution. Experimental results show that the improved algorithm is efficient for rotated and noisy images. Contrasted with the original phase correlation based algorithm, equal error ration (EER) is reduced 40.8%.
展开▼
