Comparative analysis of different image denoising methods in spatial and transform domain

Digital Images are the best source for humans to see, visualize, think, extract information and make conclusions. However during the acquisition of images, noise superimposes on the images and reduces the information and detail of the images. In order to restore the details of the images, noise must be reduced from the images. This requirement places the image denoising amongst the fundamental and challenging fields of computer vision and image processing. In this project six fundamental techniques / algorithms of image denoising in spatial and transform domain are presented and their comparative analysis is also carried out. The noise model used in this project is Additive Gaussian noise. The algorithms are simulated on Matlab and experimental results are shown at different noise levels. The performance of each image denoising technique is measured in terms of Peak Signal to Noise Ratio (PSNR) , Mean Structural Similarity (SSIM) Metrics and visual quality. It is observed that the transform domain techniques used in this project achieved better results as compared to spatial domain techniques