K-means clustering for adaptive wavelet based image denoising

Clustering algorithms are used for systematic retrieval of data by organizing them into several clusters. K-Means is one such algorithm which partitions data into groups based on distance metric in an unsupervised way. Clustering is used to organize data for efficient retrieval. In this paper, we study Denoising of images corrupted with variable Gaussian noise spread across the images (dataset). The dataset was made by applying K-Means grouping statistical parameters of the training images which are present in wavelet domain. Adaptive Soft thresholding of noisy images is done, selecting the best parameter based on the cluster. After applying inverse wavelet transform PSNR of the denoised image is calculated. Impressive results are obtained by applying this technique.