Robust detection of singular points for fingerprint recognition

Detection of fingerprint singular points is an important and difficult task in automatic fingerprint classification and identification. In this paper, a robust method for the detection of such points is presented. An improved Poincare index, the orientation variation measure and the use of gray value variance, and combination of them are proposed for fingerprint singular point detection and positioning. Experimental results on fingerprint images and comparison with traditional methods show less computational complexity and more robustness of the method proposed.