Blind Noise Level Detection
In this paper we present a new fully automatic algorithm for blind noise level evaluation based on natural image statistics (NIS). Natural images are unprocessed reproductions of a natural scene observed by a human. During its evolution, the Human Visual System has been adjusted to the information encoded in natural images, making images interpreted best by a human when they fit NIS. The main requirement of such statistics is their striking regularity. Unfortunately, most computer images suffer from various artifacts, such as noise, that distort this regularity. Our contribution is applying the statistical behaviors for noise level evaluation. As most denoising algorithms require the user to specify the noise level automatization of the process makes it more usable and user independent. We compare the quality of our results to other algorithms.
Keywordsnoise level detection data restoration computer graphics
Unable to display preview. Download preview PDF.
- 2.Simoncelli, E.P.: Statistical Modeling of Photographic Images. In: Handbook of Video and Image Processing, 2nd edn. Alan Bovik, Academic Press (2005)Google Scholar
- 4.Liu, C., Freeman W.T., Szeliski R., Kang, S.B.: Noise Estimation from a Single Image. In: IEEE Conf. on Computer Vision and Pattern Recognition (2006)Google Scholar
- 5.Kaur, L., Gupta, S., Chauhan, R.C.: Image Denoising using Wavelet Thresholding. In: ICVGIP (2002)Google Scholar
- 7.Tadmor, E., Nezzar, S., Vese, L.: A multiscale image representation using hierarchical (BV,L2) decompositions. Multiscale Model. Simul. (2), 554–579 (2004)Google Scholar
- 10.Hasinoff, S.W., William, F.D., Freeman, T.: Noise-Optimal Capture for High Dynamic Range PhotographyGoogle Scholar
- 11.Coifman, R.R., Donoho, D.: Translation-invariant de-noising. Wavelets and Statistics, pp. 125–150. Springer (1995)Google Scholar
- 15.Simoncelli, E.: Statistical Models for Images: Compression, Restoration and Synthesis. In: 31st Asilomar Conf. on Signals, Systems and Computers, pp. 673–678. IEEE Computer Society (1997)Google Scholar