An approach for multimodal biometric fusion under the missing data scenario

While biometric fusion is a well-studied problem, most of fusion schemes cannot account for missing data (incomplete score lists), that is commonly encountered in large-scale multibiometric identification systems. In this paper, we present a new approach, where the RIBG (Robust Imputation Based on Group method of data handling) is used for handling the missing data. Since this scheme can be followed by a standard fusion scheme designed for complete data, we propose a Bees Algorithm based Weighted Sum Method (BASM) to find the optimal parameters to fuse the information given by individual matcher at match score level. The proposed method tested on the NIST multimodal database achieves 94.32% rank-1 recognition rate, even when the missing rate is set to 25%, which is overall superior to traditional approaches such as majority voting.