An Improved Spatially Selective Filter using Undecimated Wavelet Transform

This paper proposes a novel wavelet filtering method by spatial correlation thresholding based on undecimated wavelet transform (UDWT). We multiply two adjacent wavelet scales with some relative translations to line up the edges. A new threshold is presented applying to spatial correlation in our proposed scheme. Finally, we extract edge information from larger scale to smaller one only at the positions which have been extracted as edges. The simulation results indicate that our algorithm can keep edges well while removing noise greatly. It is an effective noise filtration technique especially for the signals which have abundant edge and detail information. Furthermore, it can be easily extended to edge detection, fault diagnosis, image enhancement and other applications.