Search Strategies for Image Multi-Distortion Estimation

André-Louis Caron (1) , Pierre-Marc Jodoin (1) , Christophe Charrier (2, 1)

1 MOIVRE Centre
2 Equipe Image - Laboratoire GREYC - UMR6072

In this paper, we present a method for estimating the amount of Gaussian noise and Gaussian blur in a distorted image. Our method is based on the MS-SSIM framework which, although designed to measure image quality, is used to estimate the amount of blur and noise in a degraded image given a reference image. Various search strategies such as Newton, Simplex, and brute force search are presented and rigorously compared. Based on quantitative obtained results, we show that the amount of blur and noise in a distorted image can be recovered up to 0.95% and 5.40% of accuracy respectively. To our knowledge, such precision has never been achieved before.

Domaines: Traitement des images [eess.IV]

André-Louis Caron, Pierre-Marc Jodoin, Christophe Charrier. Search Strategies for Image Multi-Distortion Estimation. IEEE International Conference on Pattern Recognition (ICPR), Aug 2010, Istanbul, Turkey. pp.2824 - 2827, ⟨10.1109/ICPR.2010.692⟩. ⟨hal-00808379⟩