A unified distance measurement for orientation coding in palmprint verification

Orientation coding based palmprint verification methods, such as competitive code, palmprint orientation code and robust line orientation code, are state-of-the-art verification algorithms with fast matching speeds. Orientation code makes use of two types of distance measure, SUM_XOR (angular distance) and OR_XOR (Hamming distance), yet little is known about the similarities and differences between these two distance measures. In this paper, we propose a unified distance measure, of which both SUM_XOR and OR_XOR can be regarded as special cases, and provide some principles for determining the parameters of the unified distance. Experimental results show that, using the same feature extraction and coding methods, the unified distance measure gets lower equal error rates than the original distance measures.