A multimodal biometric user Authentication framework based on Face recognition and signature Signals using integrated Classification techniques

Abstract

In biometrics, choosing of right modality is a challenging task for the recognition of a person. Due to the advantage of widely accepted identification, face recognition systems, signature-based biometric modality is selected as a significant pattern as compared with other modalities. On the other hand, multi-biometrics aims to improve the quality of recognition over an individual method by combining the results of multiple features, sensors, or algorithms. Different Face and signature sequences of the same subject may contain variations in resolution, illumination, pose, facial expressions and signing position. These variations add to the difficulties in planning a viable multimodal-based face and signature recognition algorithm. In this Research work, an efficient Synthesis score fusion-based MNN, Legion feature Neural Network (LFNN) and Self-Organizing Map (SOM) with a neural network classifiers-based face Signature recognition system for Authentication is proposed to reduce the computational complexity of the existing method. Here, data glove signaling means of signing process are taken into account to do signature verification system. Hence the proposed works have used face and data glove signal patterns to features-level fusion for the verification system. The projected modeling is applied in the Matlab 2013a environment for different test conditions compared with the conventional method of regulating the precision, Accuracy, Recall, F-measure, Sensitivity, specificity, and time complexity parameters the advantages and robustness of the proposed design newline