Analyzing the impact of non-biometric information on multiclassifier processing for signature recognition applications

This paper describes an empirical study of ways in which the integration of biometric and non-biometric information can be exploited in signature recognition applications. The analysis investigates the impact of such an approach both on the performance enhancement achievable and the improvements attainable in relation to the flexibility with which such a system can be deployed. We present wide-ranging experimental results and an assessment of the practical implications of our approach.