A Biometric Fusion Based on Face and Fingerprint Recognition using ANN

Abstract

Biometric systems are used for identifying and recognizing individual characteristics on the basis of biological or behavioral features. In the research work, a biometric fusion system based on fingerprint and face using the artificial intelligence technique is proposed. To achieve better accuracy of the biometric fusion system, the uniqueness of feature is significant. To find out the unique feature set from the data, we have used different feature extraction algorithm in the proposed biometric fusion system. Initially, pre-processing has been applied on the test images which is used to remove the unwanted data from the uploaded image and return an appropriate data for further process. In the fingerprint part, minutia extraction is used as a feature of fingerprint whereas Extended Local Binary pattern (ELBP) is used for extracting features of face and creates a pattern of face features. To create a unique feature set, optimization algorithm is needed and we have used genetic algorithm as a feature optimization technique. In the proposed fusion system, ANN is used to classify the test data according to the trained ANN structure with optimized feature data of fingerprint and face. To check the efficiency of proposed fusion system, we have calculated the performance parameters like FAR, FRR and Accuracy. From the analysis of proposed fusion system, we have observed that the accuracy of the proposed work is better than the previous ones and it is more than the 94%. To design a proposed biometric fusion system, image processing toolbox is used under the MATLAB environment.