FACE AND FINGERPRINT FUSION SYSTEM FORIDENTITY AUTHENTICATION USING FUSIONCLASSIFIERS

摘要

In this work, we propose a feature level fusion and decision level fusion of face and fingerprintfor designing a multimodal biometric system. Initially, Gabor and Scale Invariant FeatureTransform features are extracted for both offline face and fingerprint of a person and studied theidentification accuracy. Later the fusion of the biometric traits is recommended at feature levelusing all possible combinations of feature vectors. The possible combination of features is fedinto fusion classifier of K-Nearest Neighbour(KNN), Support Vector Machine (SVM), NavieBayes(NB) and Radial Basis Function(RBF). The best combination of feature vectors and fusionclassifiers is identified for the proposed multimodal biometric system. Experiments areconducted on Face database and fingerprint database to assess the actualadvantage of the fusionof these biometric traits, in comparison to the unimodal biometric system. Experimental resultsreveal that fusion combination outperforms individual.