Noisy Image Compressive Sensing Based on Nonlinear Diffusion Filter

Abstract:

In the theory of compreesive sensing, the small amount of signal values can be reconstructed when signal is sparse or compressible.But the reconstruction of noisy image isnt very satisfied.In order to improve the quality of reconstruction image,the nonlinear diffusion filter is used in this paper.From the experiment results,the images reconstructed after nonlinear diffusion filter are better,and the value of PSNR is improved.

• Periodical: Applied Mechanics and Materials (Volume 510)
• Pages: 278-282
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.510.278
• Online since: February 2014
• Authors: Ying Zhu*, Yong Xing Jia, Yuan Wang, Chuan Zhen Rong, Yu Yang
• Keywords: Gaussian Filter, Noisy Image, Nonlinear Diffusion Filter, Signal Reconstruction

References

[1] Perona P, Mailk J. Scale space and edge detection using anisotropic diffusion[J]. IEEE PAMI, 12, 629-639, (1990).

DOI: 10.1109/34.56205

[2] Joel A. Tropp, Signal Recovery From Random Measurements Via orthogonal Matching Pursuit. IEEE TRANSACTION ON INFORMATION, VOL. 53, 2007. 12.

DOI: 10.1109/tit.2007.909108

[3] Patrick S. Huggins, Steven W. Zucker, Greedy Basis Pursuit. IEEE TRANSACTION ON INFORMATION, VOL. 55, 2007. 07.

[4] Wei Dai, Subspace Pursuit for Compressive Sensing Signal Reconstruction, IEEE TRANSACTION ON INFORMATION THEORY, VOL. 55, 2009. 05.

[5] Dong Sik Kim, Quantization Constrained Convex Optimation for the Compressive Sensing Reconstructions. ICASSP (2010).