ON IMPROVING THE NOISE REDUCTION ALGORITHMS USING IMAGE SEGMENTATION

This paper discusses how a partial segmentation of an image can be used to enhance the performance of existing wavelet noise reduction algorithms. It demonstrates that the spatial decomposition, provided by the segmentation, complements the frequency based decomposition of the wavelet trans-form, leading to better localisation of objects and allowing improvement of the noise reduction schemes. 1. BACKGROUND Image noise reduction is a low-level tool utilised in a variety of ways within range of image processing tasks, on of the most obvious examples being the enhancement of images for archiving purposes. A great many noise reduction schemes have been proposed. One of the most popular and powerful classes of methods is based on wavelet decomposi