MULTIMODAL BIOMETRIC SCORE LEVEL FUSION USING ADVANCED OPTIMIZED FUZZY INFERENCE SYSTEM

The biometric system's primary objective is to automatically differentiate between individuals and to secure records. Also, it defends access to services against unauthorised users. Here it has been developed a biometric authentication framework which is a valid alternative to traditional methods. These demands are difficult for the single modality of biometric systems today and the solution is to integrate additional information sources to enhance the decision-making process. A multi-biometric system brings together information from several biometric characteristics, algorithms, sensors and other components in a decision on recognition. Research and marketing activities in this area have experienced exponential growth in recent years, and this trend is expected to continue. In this article, we propose a novel Advanced Optimized Fuzzy Interference System (AOFIS) approach to fusing the iris and fingerprint at score level for multimodal biometric authentication fusion. We mainly explore indepth fusion and their potential application as biometric identifications of iris and fingerprint. In score levels, individual comparative scores obtained from the iris and fingerprints are combined to identify the user as genuine or imposter based on scores of fusion methods. In this study, we used Min-Max normalization techniques for Gabor-HOG as already existing and compared this with our proposed AOFIS method. The database templates and input data are compared by the FAR, FRR and accuracy error rates parameters. The FAR, FRR and ACCURACY are compared with various threshold levels. It had obtained minimal False Acceptance Rate (FAR) & False Rejection Rate (FRR) for AOFIS in experimental analysis and high AccuracyRate while comparing with Gabor-HOG. Fused scores are thus used to classify an unknown user as the genuine or impostor. It can be used for multipurpose purposes such as authentication and attendance in any kind of company as banking, government subsidies, etc.