Estimation of the amount of noise in digital images using finite differences

Abstract

The paper presents a novel method for estimating the variance of noise in a digital image. This method is based on filtering corrupted image using finite differences. As a result an original signal is removed while noise remains so its variance can be calculated. In the paper several differential filter’s masks of various sizes were derived. The proposed method was verified on test images corrupted by the additive Gaussian distributed noise. The method was compared with several previously published estimation methods. It was shown that the novel method performs well for a large range of noise variance values.