A realization of edge-preserving smoothing filters using layered neural networks

A realization of the edge-preserving smoothing filter suitable for a mixed Gaussian and impulsive noise environment is proposed. This filter consists of a layered neural network and 5-points median filter. By using the layered neural network, parameters of the proposed filter can adapt themselves to various noisy environments through the learning of a training image. The training method of the parameters of the response function is also proposed. These parameters have an important effect on the performance of the proposed filters. An example is shown to illustrate the application of the proposed filter