Digital images are corrupted by noise introduced therein by factors of sorts viz. nature of acquisition devices and processes- quantization, compression and transmission. This degradation of images aesthetically affects the human perception and concomitantly the processes of feature recognition, segmentation, and edge detection to name a few. For accurate interpretation of these images, image denoising techniques are banked upon. The purpose of denoising remains estimation of the original image from its degraded version alongwith preserving of complex structures of images, such as, edges and textures. The isotropic basis elements of wavelets fail to capture the line singularities, curve singularities and texture. Many multiresolution transforms are available that have been successful in representing specific regions of an image, but they suffer from presence of artifacts in areas not belonging to their domain. In order to denoise an image contaminated with additive white Gaussian noise, many hybrid methods have been introduced in this book that combine the features of more than one transform that can better preserve the structural details.