Denoising an Image by Denoising Its Curvature Image

In this article we argue that when an image is corrupted by additive noise, its curvature image is less affected by it; i.e., the peak signal-to-noise ratio of the curvature image is larger. We speculate that, given a denoising method, we may obtain better results by applying it to the curvature image and then reconstructing from it a clean image, rather than denoising the original image directly. Numerical experiments confirm this for several PDE-based and patch-based denoising algorithms.