Quality measures in unimodal and multimodal biometric verification

Real-world deployment of unimodal and multimodal biometric systems often have to contend with degraded signal quality and erratic behaviour of the biometric data being modelled. We review approaches that have been used to extract additional information about the biometric data that can then be used to improve performance in degraded conditions, with a special emphasis on speech (where we present new approaches for signal quality estimation in biometric verification), face, fingerprint, and signature modalities. We also present approaches that do not depend on specific modalities, including new user-model based quality measure. We show how this information can be used in a unimodal and multimodal context, and we perform objective evaluation of quality measures on multimodal benchmarking databases.