Total variation remote sensing image restoration based on split Bregman method

被引：0

作者：

Xu, Mengxi ^{[1
,2
]}

Sun, Quansen ^{[2
]}

Zheng, Shengnan ^{[3
]}

Huang, Chenrong ^{[3
]}

机构：

[1] Nanjing Institute of Technology, China

[2] Nanjing University of Science and Technology, China

[3] Nanjing Institute of Technology, China

来源：

International Journal of Digital Content Technology and its Applications | 2012年 / 6卷 / 20期

关键词：

Bregman; Image restoration; Noise model; Remote sensing; Total variation;

D O I：

10.4156/jdcta.vol6.issue20.8

中图分类号：

学科分类号：

摘要：

During the generation of the remote sensing image, there is severe image degradation, because of the sensor limit and atmospheric turbulence. Compared with common image, the remote sensing image should have richer detail after restoration. However, we find it difficult to use traditional image restoration methods, such as Wiener filtering and Richardson-Lucy (RL), to effectively suppress noise and preserve details. Furthermore, the default Gaussian noise model is not accurate. In the paper, we restore the remote sensing image based on total variation (TV) regularization, which is able to remove different noise and keep details. To minimize the TV regularization functional, we use the split Bregman method. This introduces auxiliary variables to transform the problem into three simple subproblems, which reduces the computational complexity. It is shown in experiments that the proposed method can restore the remote sensing image with different noise and effectively preserve details. Compared with Peak value Signal-to-Noise Ratio (PSNR) of FTVd method, that of proposed is improved by about 0. 3db. And it uses less time than FTVd method. It also performs well for other images.

引用

页码：71 / 79

页数：8

共 19 条

[1]

Wang H.B., Zheng S.N., Et al., AN APPROACH FOR TARGET DETECTION AND EXTRACTION BASED ON BIOLOGICAL VISION, Intelligent Automation and Soft Computing, 17, 7, pp. 909-921, (2011)

[2]

Rudin L., Osher S., Fatemi E., Nonlinear total variation based noise removal algorithm, Physical D: Nolinear Phenomena, 60, 4, pp. 259-268, (1992)

[3]

Krishnan D., Fergus R., Fast image deconvolution using hyper-laplacian priors, NIPS, 22, pp. 1-9, (2009)

[4]

Xu L., Jia J.Y., Two-phase Kernel Estimation For Robust Motion Deblurring, pp. 157-170, (2010)

[5]

Yuan L., Sun J., Quan L., Shum H.Y., Image deblurring with blurred/noisy image pairs, ACM Trans. On Graph. SIGGRSPH, 26, pp. 1-10, (2007)

[6]

Dey N., Blance-Feraud L., Zimmer C., Et al., Richardson-Lucy algorithm with total variation regularization for 3-D confocal microscope deconvolution, Microscopy Research Technique, 26, 69, pp. 260-266, (2006)

[7]

Ding X.F., Xu L.Z., Et al., Stereo depth estimation under different camera calibration and alignment errors, Applied Optics, 50, 10, pp. 1289-1301, (2011)

[8]

Chan T.F., Shen J., Image Processing and Analysis: Variational, PDE, Wavelet, and Stochastic Methods, (2005)

[9]

Tadmor E., Nezzar S., Vese L., A multiscale image representation using hierarchical (BV, L2) decompositions, Multiscale Modeling & Simulation, 2, 4, pp. 554-579, (2004)

[10]

Chambolle A., An algorithm for total variation minimization and applications, Journal of Mathematical Imaging and Vision, 20, pp. 1-2, (2004)

← 1 2 →