Double Half-Orientation Code and Nonlinear Matching Scheme for Palmprint Recognition

Abstract

The algorithm of palmprint recognition based on direction coding is an important algorithm supplement of identity authentication at this stage. In this paper, a palmprint recognition algorithm based on bi-directional coding and nonlinear matching is proposed. By modifying the original Gabor filter, this paper defines a group of “half-Gabor” filters, and adopts the “half-Gabor” filters so as to extract the double half-orientation code of the palmprint. Experimental results on the three categories of palmprint database show the high precision performance of the method proposed in this paper compared with the prior art method. The double-half Direction Codes are matched by Non-linear Matching Method. Moreover, the method is compared with Hamming distance matching method, which proves the superiority of this method in precision. The results show that the two algorithms are effective and accurate.