Modular Neural Networks with granular fuzzy integration for human recognition

In this paper a new model of a Modular Neural Network (MNN) with fuzzy integration using a granular approach is proposed. The main goal of the proposed approach is to obtain an optimal number of sub modules and optimal percentage of data for training in the MNN. The model was applied to pattern recognition based on the ear and voice biometrics. The proposed method is able to divide the data automatically into sub modules, to work with a percentage of images and select which are the optimal images to be used for training. Also a Hierarchical Genetic Algorithm (HGA) for MNN optimization is proposed. Finally, fuzzy logic as a method for MNN response integration of these biometrics measures is used.