A Survey on Various Image De-noising Methods

Abstract

Image de-noising is a classical yet fundamental problem in low level vision, as well as an ideal test bed to evaluate various statistical image modeling methods. The restoration of a blurry or noisy image is commonly performed with a maximum a posteriori probability (MAP) estimator, which maximizes a posterior probability to reconstruct a clean image from a degraded image. A MAP estimator, when used with a sparse gradient image prior, reconstructs piecewise smooth images and typically removes textures that are important for visual realism. One of the most challenging problems in image de-noising is how to preserve the fine scale texture structures while removing noise. Various natural image priors, such as gradient based prior, nonlocal self-similarity prior, and sparsity prior, have been extensively exploited for noise removal. The de-noising algorithms based on these priors, however, tend to smooth the detailed image textures, degrading the image visual quality. To address this problem, we propose a texture enhanced image de-noising (TEID) method by enforcing the gradient distribution of the de-noised image to be close to the estimated gradient distribution of the original image. Another method is an alternative de-convolution method called iterative distribution reweighting (IDR) which imposes a global constraint on gradients so that are constructed image should have a gradient distribution similar to a reference distribution.

Keywords: types of noise, image de-noising, image deblurring, comparison, proposed algorithm