Median filtering in the wavelet domain-in image segmentation

In this paper we propose a method of applying median filter in the wavelet domain in image segmentation when the image consisting of regions hard to be segmented by spatial domain median filter. The method transforms an image into the wavelet domain and then iteratively applies median filter in the wavelet domain and finally transforms result into the spatial domain. The advantage of wavelet domain median filter is that in the wavelet domain, probabilities of encountering root images are spread over sub-band images and therefore median filter is unlikely to encounter root images at early stage of iterations and can generate better result as iteration increases. Better performance are obtained in segmenting images having a Gaussian noise patter, especially when regions are close to each other in their mean values. Results from computer simulation are used to demonstrate superiority of the method.