Image denoising is a noteworthy range of examination because of the need of clear pictures in the field of straightforward photography, therapeutic picture distinguishment, Human Shape digitization and so forth. There are a few procedure has been connected when we will work with pictures like getting, changing, coding and disentangling. So the shots of clamor to be embedded in the genuine pictures are high. In this paper we proposed an efficient denoising based on linear threshold expansion. This method work on domain/edge basis transformation. So this technique has improved capability of retrieving the images. The higher PSNR and lower RMSE values show the effectiveness of our approach.