Complete Binary Representation for 3D Palmprint Recognition
摘要
In this chapter, we propose a complete binary representation (CBR) for the 3D palmprint multiple dimensional feature representation and recognition. First, we propose a multiple orientation binary representation (MOBR) to extract 2D gray-level features of 3D palmprint. Then, we propose a novel and effective compact surface type (CST) to characterize 3D palmprint surface feature and design a CST binary representation (CSTBR) to capture the surface consistency within a local patch. Finally, we develop a CBR method by integrating the MOBR and CSTBR, which can effectively represent both 2D orientation-level and 3D surface-level features of 3D palmprint images. We conduct extensive inter-comparison experiments to demonstrate the effectiveness of the proposed CBP method on a benchmark 3D palmprint database. Also, we carry out multiple intra-comparisons to validate the superiority of the MOBR and CST-based representations.