Feature Extraction of Dorsal Hand Vein Pattern Using a Fast Modified PCA Algorithm Based On Cholesky Decomposition and Lanczos Technique

Abstract

Dorsal hand vein pattern is an emerging biometric which is attracting the attention of researchers, of late. Research is being carried out on existing techniques in the hope of improving them or finding more efficient ones. In this work, Principle Component Analysis (PCA) , which is a successful method, originally applied on face biometric is being modified using Cholesky decomposition and Lanczos algorithm to extract the dorsal hand vein features. This modified technique decreases the number of computation and hence decreases the processing time. The eigenveins were successfully computed and projected onto the vein space. The system was tested on a database of 200 images and using a threshold value of 0.9 to obtain the False Acceptance Rate (FAR) and False Rejection Rate (FRR). This modified algorithm is desirable when developing biometric security system since it significantly decreases the matching time Keywords—Dorsal hand vein pattern, PCA, Cholesky Decomposition, Lanczos algorithm.

2 Figures and Tables

Cite this paper

@inproceedings{Heenaye2010FeatureEO, title={Feature Extraction of Dorsal Hand Vein Pattern Using a Fast Modified PCA Algorithm Based On Cholesky Decomposition and Lanczos Technique}, author={Maleika Heenaye and M. Abdesh S. K. Khan and N. M. Khan and Raja K.Subramanian}, year={2010} }