A novel approach for Online signature verification using fisher based probabilistic neural network

The rapid advancements in communication, networking and mobility have entailed an urgency to further develop basic biometric capabilities to face security challenges. Online signature authentication is increasingly gaining interest thanks to the advent of high quality signature devices. In this paper, we propose a new approach for automatic authentication using dynamic signature. The key features consist in using a powerful combination of linear discriminant analysis (LDA) and probabilistic neural network (PNN) model together with an appropriate decision making process. LDA is used to reduce the dimensionality of the feature space while maintaining discrimination between users. Based on its results, a PNN model is constructed and used for matching purposes. Then a decision making process relying on an appropriate decision rule is performed to accept or reject a claimed identity. Data sets from SVC 2004 have been used to assess the performance of the proposed system. The results show that the proposed method competes with and even outperforms existing methods.