Person recognition by fusing palmprint and palm vein images based on Laplacianpalm representation

摘要: Unimodal analysis of palmprint and palm vein has been investigated for person recognition. One the problems with unimodality is that unimodal biometric less accurate vulnerable to spoofing, as data can be imitated or forged. In this paper, we present a multimodal personal identification system using images their fusion applied at image level. The are fused by new edge-preserving contrast-enhancing wavelet method in which modified multiscale edges combined. We developed rule enhances discriminatory information images. Here, novel representation, called ''Laplacianpalm'' feature, extracted from locality preserving projections (LPP). Unlike Eigenpalm approach, finds an embedding preserves local yields space best detects essential manifold structure. compare proposed approach Fisherpalm methods on large set. Experimental results show provides better representation achieves lower error rates Furthermore, outperforms any its individual modality.