Multibiometric Identification System based on SVD and Wavelet Decomposition

Abstract

Biometric systems refer to the systems used for human recognition based on their characteristics. These systems are widely used in security institutions and access control. In this work three biometric sources were used for identification purposes. Singular value decomposition (SVD) was employed as a tool for feature extraction and artificial neural network (ANN) was used as pattern recognition for the model. High accuracy was obtained from this work with 95% recognition rate.