GwPeSOA-based MSVNN: the multimodal biometric system for futuristic security applications

Abstract

Biometric systems have gained considerable significance as they are highly employed in the security applications. Achieving human recognition is easier and cheaper and the single modality employed for the recognition faces a lot of challenges due to the environmental factors. Thus, the paper proposes a multimodal recognition system based on the Multi-Support Vector Neural Network (MSVNN). The algorithm proposed is the Glowworm Penguin Search Optimization Algorithm (GwPeSOA), which is a modification of the Glowworm Optimization Algorithm (GOA) with the Penguin Search Optimization Algorithm (PeSOA). The proposed method employs two modalities: the ear and the finger vein modalities; the features of the ear image are obtained using the proposed BiComp masking method of feature extraction, whereas the features from the finger vein are extracted using the Repeated Line Tracking (RLT) method. The features obtained are applied to the MSVNN classifier to recognize the person with good accuracy and the proposed BiComp Mask offers robust features for the extraction. The experimentation using the proposed method attained a better accuracy, specificity and sensitivity at a rate of 0.95, 0.95 and 0.9868, respectively.