作者: Daniela Sánchez , Patricia Melin , Oscar Castillo , Fevrier Valdez
DOI: 10.1007/978-3-642-37798-3_22
关键词:
摘要: In this paper a new model of Modular Neural Network (MNN) with fuzzy integration based on granular computing is proposed. The topology and parameters the MNN are optimized Hierarchical Genetic Algorithm (HGA). proposed method can divide data automatically into sub modules or granules, chooses percentage images selects which will be used for training. responses each module combined using integrator, number integrators depend granules that has at particular moment. was applied to case human recognition illustrate its applicability good results.