An Efficient Multi-algorithmic Fusion System Based on Palmprint for Personnel Identification

Abstract

This paper proposes an efficient biometric system for personnel identification of an individual. It is based on Karhunen-Loeve and Haar wavelet transform features of human hand. A low cost scanner is used to acquire image for biometric features. The palmprint features which are independent to hand translation and rotation on the scanner are extracted. Fusion at the matching score level is done by means of the dynamic weighted sum rule. The system is tested on IITK database of 200 images and PolyU Database of 7751 images. The system is found to have 94.80% accuracy with 4.73% FAR and 5.66% FRR.