Fusion based denoise-engine for underwater images using curvelet transform

A novel image de-noising algorithm is proposed, which utilizes the effective benefits of curvelet and fusion techniques to improve the quality of underwater images. With the help of this algorithm, the de-noised image is obtained by merging two intermediate images of the same. In this framework, the input noise image is simultaneously fed into gaussian filter and curvelet transform. The first output image is obtained by filtering the image with Gaussian. The second output image is obtained by denoising it with curvelet transform. Experimental results show that the PSNR and MSE ratio is substantially improved by the proposed method. This method is efficient at denoising while saving edges and textures in underwater images.