Sparse Signal Detection and Fingerprint Feature Recognition Based on Fast 2D DFRFT

Abstract

In this paper, a two-dimensional fractional Fourier transform (2D FRFT) based fingerprint feature extraction scheme is proposed, based on the fact that the fingerprint image can be approximated as two-dimensional chirp signals. And the sparse 2D fractional Fourier transform (STDFRFT) algorithm is proposed to achieve efficient computation of 2D FRFT. The effectiveness of the proposed STDFRFT algorithm is reflected in the simulations and applications of fractional domain sparse random signal detection, convergence analysis, two-dimensional chirp signal detection and fingerprint feature recognition.