Removing Mixture of Gaussian and Impulse Noise by Patch-Based Weighted Means
- 345 Downloads
We first establish a law of large numbers and a convergence theorem in distribution to show the rate of convergence of the non-local means filter for removing Gaussian noise. Based on the convergence theorems, we propose a patch-based weighted means filter for removing an impulse noise and its mixture with a Gaussian noise by combining the essential idea of the trilateral filter and that of the non-local means filter. Experiments show that our filter is competitive compared to recently proposed methods. We also introduce the notion of degree of similarity to measure the impact of the similarity among patches on the non-local means filter for removing a Gaussian noise, as well as on our new filter for removing an impulse noise or a mixed noise. Using again the convergence theorem in distribution, together with the notion of degree of similarity, we obtain an estimation for the PSNR value of the denoised image by the non-local means filter or by the new proposed filter, which is close to the real PSNR value.
KeywordsGaussian noise Impulse noise Mixed noise Trilateral filter Non-local means filter Convergence theorems Degree of similarity Estimation of PSNR
The work has been supported by the Fundamental Research Funds for the Central Universities in China (No.N130323016), the Research Funds of Northeastern University at Qinhuangdao (No. XNB201312, China), the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry(48-2), the National Natural Science Foundation of China (Grant Nos. 11171044 and 11401590), and the Science and Technology Research Program of Zhongshan (Grant No. 20123A351, China). The authors are very grateful to the reviewers for their valuable remarks and comments which led to a significant improvement of the manuscript. They are also grateful to Prof. Raymond H. Chan and Dr. Yiqiu Dong for kindly providing the code of ROLD-EPR.
- 2.Bilcu, R.C., Vehvilainen, M.: Fast nonlocal means for image denoising. In: Proceedings of SPIE, Digital Photography III 6502, 65020R (2007)Google Scholar
- 5.Castleman, K.: Digital Image Processing. Prentice Hall Press, Upper Saddle River (1996)Google Scholar
- 12.Delon, J., Desolneux, A.: A patch-based approach for random-valued impulse noise removal. In: 2012 IEEE international conference on acoustics, speech and signal processing (ICASSP), pp. 1093–1096. IEEE (2012)Google Scholar
- 20.Hu, H., Froment, J.: Nonlocal total variation for image denoising. In: Symposium on Photonics and Optoelectronics (SOPO), pp. 1–4. IEEE (2012)Google Scholar
- 21.Hu, H., Li, B., Liu, Q.: Non-local filter for removing a mixture of gaussian and impulse noises. In: Csurka, G., Braz, J. (eds.) VISAPP, vol. 1, pp. 145–150. SciTePress (2012)Google Scholar
- 28.Muresan, D., Parks, T.: Adaptive principal components and image denoising. In: IEEE International Conference on image processing (2003)Google Scholar
- 30.Pratt, W.: Median filtering. Semiannual Report, Image Processing Institute, University of Southern California, Los Angeles, Tech. Rep pp. 116–123 (1975)Google Scholar
- 34.Tomasi, C., Manduchi, R.: Bilateral filtering for gray and color images. In: Proceedings of the international confernce on computer vision, pp. 839–846. IEEE (1998)Google Scholar
- 39.Xu, H., Xu, J., Wu, F.: On the biased estimation of nonlocal means filter. In: International conference on multimedia computing and systems/international conference on multimedia and expo, pp. 1149–1152 (2008)Google Scholar
- 40.Yang, J.X., Wu, H.R.: Mixed gaussian and uniform impulse noise analysis using robust estimation for digital images. In: Proceedings of the 16th international conference on digital signal processing, pp. 468–472 (2009)Google Scholar