A Fuzzy Evidential Reasoning Data Fusion Approach with Uncertainty Evaluation for Robust Pattern Classification

Abstract

This paper presents a data fusion approach for pattern classification, based on the fuzzy evidential reasoning technique. First, a new fuzzy evidence structure model is introduced to formulate probabilistic evidence and fuzzy evidence in a unified manner. A generalized Dempster's rule is then used to combine fuzzy evidence structures associated with multiple information sources. Finally, an effective decision rule is developed to take into account uncertainty, quantified by Shannon entropy and fuzzy entropy, of probabilistic evidence and fuzzy evidence, to deal with conflict and to achieve robust decisions. To demonstrate the effectiveness of this approach, we apply it to classify multimodality human brain MR images in a supervised manner. The proposed approach outperforms both the traditional evidential reasoning technique and the fuzzy reasoning technique, in terms of robustness and classification accuracy.