Experimental Evaluation of Matching-Score Normalization Techniques on Different Multimodal Biometric Systems

The goal of this work is the experimental evaluation of matching-score normalization techniques for the following three multimodal biometric systems: a verification system based on the fusion of face and palmprint principal line features, an identification system based on eigenfingers and finger-geometry, and an identification system based on eigenpalm and eigenfinger features. The following normalization techniques are tested: Bayes-based normalization and several heuristic techniques (min-max, z-score, median-MAD, double-sigmoid, tanh, and piecewise-linear). The results of evaluation are represented by means of system performance (expressed by ROC, EER and minTER) and F-statistics