Supervised texture segmentation using wavelet transform

This paper presents a supervised segmentation algorithm based on wavelet transform for the textured image of remote sensing. A discrete wavelet frame is adopted to decompose an image into multichannel images. An improved method for feature extraction is developed in this paper. It is adaptive and takes the nonstationary characteristics of noise filtering into account. Further, this method incorporates contextual/spatial information among feature images to reduce variability of texture feature estimates while retaining the accuracy of region boundaries. In the stage of segmentation, the estimated feature vector of each pixel is sent into a Bayes classifier to make an initial probabilistic labeling. To obtain a more accurate result of segmentation, a probabilistic relaxation method is used to introduce the spatial constraints into the segmentation algorithm. Finally, the performance of the proposed segmentation algorithm is demonstrated on a variety of images including remote sensing images.