Fuzzy Exponential Recurrent Neural Networks for Gray-scale Image Retrieval ?
Marcos Eduardo Valle · 2014
Associative memories (AMs) are mathematical models inspired by the human brain ability to store and recall information. This paper introduces the fuzzy exponential recurrent neural networks (FERNNs), which can implement an AM for the storage and recall of fuzzy sets. The novel models are obtained by modifying the multivalued exponential recurrent neural network of Chiueh and Tsai. Briefly, a FERNN defines recursively a sequence of fuzzy sets obtained by averaging the stored fuzzy sets weighted by an exponential of a fuzzy comparison measure between the current fuzzy set and the stored items. Computational exper- iments reveal that FERNNs can be effectively used for the retrieval of gray-scale images corrupted by either Gaussian noise or salt-and-pepper noise.