Neo Fuzzy Neuron Filters and Their Applications to Image Signal Preprocessing
Noriaki Suetake, Takeshi Yamakawa · IEEJ Transactions on Electronics Information and Systems · 1999
We propose novel nonlinear filters which are the extensions of a linear FIR filter and an order statistic (OS) filter by employing the neo fuzzy neuron (NFN) model. We also propose the hybrid type nonlinear filter aiming at elimination of a Gaussian noise and an impulsive noise at the same time, and high restoration of the signal, simultaneously. The proposed filters are synthesized by a learning method which guarantees optimal design caused by employing the NFN model. Moreover, the filters are effective not only for noise elimination but also for sharpening, edge extraction, and other various applications, because their functions are determined by the pairs of target and input signals in the training. The effectiveness and validity of the proposed filters are verified by applying them to the preprocessing of the image signals.