Fuzzy Expert System for Automatic Wavelet Shrinkage Procedure Selection for Noise Suppression

Signal processing is an indispensable issue of several technical areas. Wavelet shrinkage, i.e. thresholding in the wavelet coefficient domain, has been successfully used for signal and image noise removal problems. Although, the selection of the suitable wavelet threshold procedure is still a challenging task, because the applied method has significant impact on the result. Furthermore, the specific choice of wavelet, decomposition level and threshold rule, etc., allows a wide variability of the shrinkage method. This paper presents a new supervisory fuzzy expert system for automatic wavelet shrinkage method selection for noise suppression of unknown signals. Simulation results show efficient performance of the system.