Issue No. 04 - October-December (2007 vol. 6)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MPRV.2007.78
Maylor K.H. Leung , Nanyang Technological University
A.C.M. Fong , Nanyang Technological University
Siu Cheung Hui , Nanyang Technological University
Access security is an important aspect of pervasive computing systems. It offers the system developer and end users a certain degree of trust in the use of shared computing resources. Biometrics offers many advantages over the username-plus-password approach for access security control. Palmprint-based personal identification has seen less research than identification using fingerprints, irises, and faces. A proposed personal verification system employs palmprint images stored as 8-bit grayscale TIFF files. The system components include mechanisms for locating and aligning, extracting, and matching palmprints. The system first detects a human hand's interfinger key points and locates the palmprint on the basis of those points. Next, the system generates a line-edge-map representation of palmprint features. Finally, it performs feature matching based on the <it>line segment Hausdorff distance</it>. The LHD matching score provides the basis for verification decisions. Experimental results demonstrate that the key-point-detection technique is effective and that the palmprint line edge map has high discriminative power. This article is part of a special issue on security and privacy.
biometrics, palmprint, authentication, personal identification, security access control
S. C. Hui, A. Fong and M. K. Leung, "Palmprint Verification for Controlling Access to Shared Computing Resources," in IEEE Pervasive Computing, vol. 6, no. , pp. 40-47, 2007.