This article addresses the structure and properties of a new class of nonlinear adaptive filters called generalized adaptive neural filters (GANF's). Various properties, such as an upper bound of the mean absolute error of the filters, are analytically derived. Experimental results are presented to demonstrate the performance of the filters for signal and image enhancement. It is shown that GANF's not only extend the class of stack filters, but also have better performance in noise suppression.
All Science Journal Classification (ASJC) codes
- Computer Science Applications
- Computer Networks and Communications
- Artificial Intelligence