SAR image iterative denoising algorithm based on non-local adaptive dictionary

doi:10.3969/j.issn.1000-1565.2018.03.012

Abstract

Abstract: In order to make better use of the information carried by SAR images, the suppression of speckle noise has become one of the hot topic around the world. Combining sparse representation theory and non local self similarity theory of images, a SAR image iterative denoising algorithm based on non-local block matching and adaptive dictionary K-SVD is proposed in this paper. Firstly, the results were matched using non local block matching algorithm of the previous iteration in each iteration, then each group of similar block adaptive dictionary update, and image block replacement dictionary is used to improve the efficiency of dictionary training, and finally the effect of SAR image denoising is achieved by K-SVD iteration. The results show that the algorithm has better denoising ability, and can better preserve the details- DOI:10.3969/j.issn.1000-1565.2018.03.012基于非局部自适应字典的SAR图像迭代去噪算法庞姣^1,2,张世琪³,刘帅奇^1,2(1.河北大学电子信息工程学院,河北保定 071002;2.河北省机器视觉工程技术研究中心,河北保定 071002;3. 北京理工大学信息与电子学院,北京 100081)摘要:为了更好地利用SAR图像携带的信息,相干斑噪声的抑制成为各国学者研究的热点之一. 结合稀疏表示理论和图像的非局部自相似理论,提出了一种基于非局部块匹配与自适应字典的K-singular value decomposition(K-SVD)的synthetic aperture radar(SAR)图像迭代去噪算法. 首先,在每次迭代中,利用非局部块匹配算法对上一次迭代的结果进行匹配分组,然后对每组相似块进行自适应字典更新,并用图像块替换字典原子来提高字典训练的效率,最后通过K-SVD的迭代实现SAR图像的去噪效果. 实验结果表明,该算法具有更好的去噪能力,能更好地保持图像的细节和纹理等有用信息. 关键词:SAR图像去噪;相似块匹配;K-SVD;迭代去噪中图分类号:TN95 文献标志码:A 文章编号:1000-1565(2018)03-0309-06SAR image iterative denoising algorithm based on non-local adaptive dictionary PANG Jiao^1,2,ZHANG Shiqi³,LIU Shuaiqi^1,2(1. College of Electronic and Informational Engineering, Hebei University, Baoding 071002, China;2. Machine Vision Engineering Research Center of Hebei Province, Baoding 071000, China;3.School of Information and Electronics,Beijing Institute of Technology,Beijing 100081,China)Abstract: In order to make better use of the information carried by SAR images, the suppression of speckle noise has become one of the hot topic around the world. Combining sparse representation theory and non local self similarity theory of images, a SAR image iterative denoising algorithm based on non-local block matching and adaptive dictionary K-SVD is proposed in this paper. Firstly, the results were matched using non local block matching algorithm of the previous iteration in each iteration, then each group of similar block adaptive dictionary update, and image block replacement dictionary is used to improve the efficiency of dictionary training, and finally the effect of SAR image denoising is achieved by K-SVD iteration. The results show that the algorithm has better denoising ability, and can better preserve the details- 收稿日期:2017-11-04 基金项目:国家自然科学基金资助项目(61572063;61401308);河北省自然科学基金资助项目(F2016201187;F2016201142);河北省高等学校科学技术研究项目(QN2016085);河北大学实验室开放项目(sy201608) 第一作者:庞姣(1979—),女,河北保定人,河北大学实验师,主要从事数字通信、信息处理、图像处理方面研究. E-mail:965961776@qq.com 通信作者:刘帅奇(1986—),男,河北石家庄人,河北大学副教授,博士,主要从事数字多维信号处理、图像处理研究. E-mail:shdkj-1918@163.com第3期庞姣等:基于非局部自适应字典的SAR图像迭代去噪算法and texture information of the image.

Key words: SAR image denoising, similar block matching, K-SVD, iterative denoising

CLC Number:

TN95

PANG Jiao,ZHANG Shiqi,LIU Shuaiqi. SAR image iterative denoising algorithm based on non-local adaptive dictionary[J]. Journal of Hebei University (Natural Science Edition), 2018, 38(3): 309-314.

References

[1] LIU S Q, SHI M Z, HU S H, et al. Synthetic aperture radar image de-noising based on Shearlet transform using the context-based model[J]. Physical Communication, 2014, 13: 221-229. DOI:10.1016/j.phycom.2014.02.002.
[2] WANG X Y, ZHANG N, ZHENG H L, et al. Extended shearlet HMT model-based image denoising using BKF distribution[J]. Journal of Mathematical Imaging and Vision, 2015, 54(3): 301-319. DOI:10.1007/s10851-015-0605-9.
[3] LU T, LI S, FANG L, et al. SAR image despeckling via structural sparse representation[J]. Sensing and Imaging, 2016, 17(1): 1-20. DOI:10.1007/s11220-015-0127-y.
[4] OZCAN C, SEN B, NAR F. Sparsity-driven despeckling for SAR images[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(1): 115-119. DOI:10.1109/lgrs.2015.2499445.
[5] OJHA C, FUSCO A, MANUNTA M. Denoising of full resolution differential SAR interferogram based on K-SVD technique[C] // Geoscience and Remote Sensing Symposium,IEEE, 2015.
[6] DEKA B, BORA P K. Removal of correlated speckle noise using sparse and overcomplete representations[J]. Biomedical Signal Processing and Control, 2013, 8(6): 520-533. DOI:10.1016/j.bspc.2013.05.003.
[7] ELAD M, AHARON M. Image denoising via sparse and redundant representations over learned dictionaries[J]. IEEE Trans Image Process, 2006, 15(12): 3736-3745. DOI:10.1109/tip.2006.881969.
[8] GOODMAN J W. Some fundamental properties of speckle[J]. Journal of the Optical Society of America, 1976, 66(11): 1145-1150. DOI:10.1364/josa.66.001145.
[9] LIU S Q, HU S H, XIAO Y, et al. Bayesian shearlet shrinkage for SAR image de-noising via sparse representation [J]. Multidimensional Systems and Signal Processing, 2013, 25(4): 683-701. DOI:10.1007/s11045-013-0225-8.
[10] OSHER S, BURGER M, GOLDFARB D, et al. An iterative regularization method for total variation-based image restoration[J]. Multiscale Modeling & Simulation, 2005, 4(2): 460-489. DOI:10.1137/040605412.
[11] DELEDALLE C, DENIS L, TUPIN F. Iterative weighted maximum likelihood denoising with probabilistic patch-based weights[J]. IEEE Trans Image Process, 2009, 16(8): 2080-2095. DOI:10.1109/tip.2009.2029593.
[12] PARRILLI S, PODERICO M, ANGELINO C V,et al. A nonlocal SAR image denoising algorithm based on LLMMSE wavelet shrinkage[J]. IEEE Trans Geosci Remote Sens, 2012, 50(2): 606-616. DOI:10.1109/tgrs.2011.2161586.
[13] XU B, CUI Y. An iterative SAR image filtering method using nonlocal sparse model[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12(8): 1635-1639. DOI:10.1109/lgrs.2015.2417551.