A New Method for Image Denoising Based on Multiresolution Technique

The need for image denoising is encountered in many practical applications. The problem with the data acquisition process, imperfect instruments and interfering natural phenomena can all degrade the data of interest. Noise can also be introduced by transmission errors. Thus it is necessary to apply an efficient denoising technique to compensate for such data corruption. The multiresolution technique i.e. curvelet transform has been employed as an efficient method in image denoising. The two phases for curvelet transform are analysis (decomposition) and synthesis (reconstruction). In the present work, a new denoising technique using hard threshold has been proposed and the results are compared with the other state of art noise reduction methods. The experimental results show that the new method is better than the other noise reduction methods in terms of quality metrics like MSE, PSNR and SSIM and reduces the Gaussian noise significantly while preserving features at the boundary of the image.