A Generic Template Update Framework: The Solution for Intra-Class Variations in Biometric Systems

It is already a consensus that intra-class variations will decline the performance of biometric systems. Many studies have explained how intra-class variations affect the biometric systems, while few approaches have proposed the effective solutions of template update. Since template update is a crucial procedure in biometric recognition systems, the problem is challenging because the newly selected templates must be representative of a large amount of intra-class variations such as posture changes and lighting conditions. To this end, existing techniques proposed to perform template update by periodically selecting new representative templates. However, all of them are designed for specific biometric traits rather a generic framework, and they are either ineffective because selected templates are less representative or prone to selection errors due to the presence of outliers. In this paper, we propose a novel generic template update framework. We use pass tables to accumulate information for outlier removal and representative templates selection, and further, propose a novel criterion to select the optimal template set according to current templates and similarity threshold. To confirm the effectiveness of our method, we carry out several experiments on synthetic and real datasets from different biometric traits. The results show that the proposed method is more robust than existing ones, and at the same time, achieves a higher accuracy according to the specified performance indicator. It also demonstrated the superiority of our method to random selection for different biometric traits.