Multimodal Biometric Reorganization System using Deep Learning Convolutional Neural Network

Biometric authentication is becoming the crucial technology in real world applications, which includes automation and security. However, the various works are implemented the unimodal biometric system, which resulted in reduced security standards. Therefore, this article is focused on implementation of multimodal biometric reorganization system (MBRS) using deep learning convolutional neural network (DLCNN) classification. Initially, images from face, fingerprint, iris datasets are applied to gaussian filter, which preprocess them and eliminates the different types of noises. Further, grey level co-occurrence matrix (GLCM) is used to extract the multimodal features. Then, Principal Component Analysis (PCA) is used to reduce the features, which selects the best features from available set. Finally, DLCNN classifier is used to perform the biometric reorganization operation from test dataset using trained features. The simulations revealed that the proposed method resulted in superior performance as compared conventional MBRS system.