Face-Iris multimodal biometric system using feedforward backpropagation neural network

Abstract

Multimodal biometric systems are used to verify or identify people by utilizing information multiple biometric modality. It combines the advantages of a unimodal biometric system to address their limitations. An efficient Face-Iris multimodal Biometric system based on artificial intelligence technique is presented in this paper. The main goal of this article is to enhance the authentication performance by fusing two biometric traits such as face and iris modalities. A feature extraction algorithm Maximally Stable Extremal Regions (MSER) along with feature optimization technique Artificial Bee Colony (ABC) is used to extract the key points and optimized these key points respectively. To detect or match face and iris Feed forward back propagation neural network (FFBPNN) is used. Evaluating overall performance of the designed modal based on accuracy, False Acceptance Rate (FAR), False Rejection Rate (FRR), Error and Receiver Operating Characteristic (ROC) analysis suggests that the proposed multimodal biometric system achieves improved results compared to existing work.