基于空间自适应非凸正则项全变差相干斑噪声抑制

李金才; 彭宇行; 朱敏; 陈鹏

doi:10.7498/aps.63.189501

摘要

全变差正则化方法是相干斑噪声抑制研究的热点. 非凸正则项能够更好地保持图像的边缘、纹理细节信息；空间自适应正则化参数可以根据像素点所在的区域，合理地控制噪声抑制程度，从而提高噪声抑制效果. 本文结合非凸正则项和空间自适应正则化参数提出了一种新的全变差相干斑噪声抑制模型，并且给出了一种模型求解的数值算法. 数值试验结果表明该模型能够获得更好的相干斑噪声抑制效果.

关键词:

Total variation is a hot point of research on speckle reduction. The nonconvex regularizer can better preserve or even enhance the information about the edges of an image. Spatially adaptive regularization parameters can reasonably control the level of speckle reduction according to the region in which the pixels are, and improve the speckle reduction effect. In this paper, we present a new total variation model for speckle reduction by integrating nonconvex regularizer and spatially adaptive regularization parameters. In order to solve the model, a new algorithm is designed based on Newton's method, augmented Lagrange multiplier, alternating direction method of multipliers, and iteratively reweighted method. The numerical examples demonstrate that the proposed model can obtain the better speckle reduction effect.

Keywords:

nonconvex regularizer /
total variation /
speckle reduction /
spatially adaptive regularization parameters

作者及机构信息

1.
国防科学技术大学计算机学院, 长沙 410073;

2.
周口职业技术学院信息工程系, 周口 466000

基金项目: 国家重点基础研究发展计划（批准号：2014CB340303）和国家高技术研究发展计划（批准号：2013AA01A213）资助的课题.

Authors and contacts

1.
School of Computer Science, National University of Defense Technology, Changsha 410073, China;

2.
Department of Information Engineering, Zhoukou Vocational and Technical College, Zhoukou 466000, China

Funds: Project supported by the National Basic Research Program of China (Grant No. 2014CB340303) and the National High Technology Research and Development Program of China (Grant No. 2013AA01A213).

参考文献

[1]	Wu Y L, Wu Z S 2014 Chin. Phys. B 23 037801
[2]	Basak K, Manjunatha M, Dutta P K 2012 Med. Biol. Eng. Comput. 50 547
[3]	Ji W J, Tong C M 2013 Chin. Phys. B 22 020301
[4]	Ma C, Gu H, Su W M, Li C Z 2014 Acta Phys. Sin. 63 028403(in Chinese)[马超, 顾红, 苏卫民, 李传中 2014 63 028403]
[5]	Argenti F, Lapini A, Bianchi T, Alparone L 2013 IEEE Geosci. Remote Sens. Magaz. 1 6
[6]	Abbott J G, Thurstone F L 1979 Ultrasonic Imaging 1 303
[7]	Chen S, Zhao H C, Zhang S N, Chen Y 2013 Acta Phys. Sin. 62 218405(in Chinese)[陈思, 赵慧昌, 张淑宁, 陈勇 2013 62 218405]
[8]	Jiang Z H, Huang S X, You X B, Xiao Y G 2014 Chin. Phys. B 23 050302
[9]	Jiang Z H, Huang S X, Shi H Q, Zhang W, Wang B 2011 Acta Phys. Sin. 60 108402(in Chinese)[姜祝辉, 黄思训, 石汉青, 张伟, 王彪 2011 60 108402]
[10]	Lee J S 1980 IEEE Trans. Pattern Anal. Mach. Intell. PAMI-2 165
[11]	Frost V S, Stiles J A, Shanmugan K S, Holtzman J C 1982 IEEE Trans. Pattern Anal. Mach. Intell. PAMI-4 157
[12]	Kuan D T, Sawchuk A A, Strand T C, Chavel P 1985 IEEE Trans. Pattern Anal. Mach. Intell. PAMI-7 165
[13]	Sun Z G, Han C Z 2010 Acta Phys. Sin. 59 3210(in Chinese)[孙增国, 韩崇昭 2010 59 3210]
[14]	Donoho D L 1995 IEEE Trans. Inform. Theory 41 613
[15]	Sveinsson J R, Benediktsson J A 2003 IEEE Trans. Geosci. Remote Sens. 41 2404
[16]	Perona P, Malik J 1990 IEEE Trans. Pattern Anal. Mach. Intell. 12 629
[17]	Yu Y J, Acton S T 2002 IEEE Trans. Image Process. 11 1260
[18]	Aja-Fernández S, Alberola-López C 2006 IEEE Trans. Image Process. 15 2694
[19]	Liu G J, Zeng X P, Tian F C, Li Z Z, Chaibou K 2009 Signal Process. 89 2233
[20]	Rudin, L, Lions P L, Ohser S 2003 Geometric Level Set Methods in Imaging, Vision, and Graphics (New York: Springer) pp103-119
[21]	Aubert G, Aujol J F 2008 SIAM J Appl. Math. 68 925
[22]	Shi J, Osher S 2008 SIAM J. Imaging Sci. 1 294
[23]	Bioucas-Dias J M, Figueiredo M A T 2009 IEEE Int. Conf. Image Processing Cairo, Egypt, November 7-10 2009 p3717
[24]	Bioucas-Dias J M, Figueiredo M A T 2010 IEEE Trans. Image Process. 19 1720
[25]	Huang Y M, Ng M K, Wen Y W 2009 SIAM J. Imaging Sci. 2 20
[26]	Steidl G, Teuber T 2010 J. Math Imaging Vis. 36 168
[27]	Yun S, Woo H 2012 IEEE Trans. Image Process. 21 2523
[28]	Woo H, Yun S 2012 IEEE Trans. Image Process. 21 1701
[29]	Han Y, Feng X C, Baciu G, Wang W W 2013 Pattern Recognition 46 989
[30]	Chen D Q, Cheng L Z 2012 IEEE Trans. Image Process. 21 1650
[31]	Wang X D, Feng X C, Huo L G 2012 Acta Automat. Sin. 38 444(in Chinese)[王旭东, 冯象初, 霍雷刚 2012 自动化学报 38 444]
[32]	Chen D Q, Cheng L Z 2014 J. Comput. Appl. Math. 257 29
[33]	Li J C, Ma Z H, Peng Y X, Huang B 2013 Acta Phys. Sin. 62 099501(in Chinese)[李金才, 马自辉, 彭宇行, 黄斌 2013 62 099501]
[34]	Sun Z G 2013 Acta Phys. Sin. 62 180701(in Chinese)[孙增国 2013 62 180701]
[35]	Nikolova M, Ng M K, Tam C P 2010 IEEE Trans. Image Process. 19 3073
[36]	Kreutz-Delgado K, Murray J F, Rao B D, Engan K, Ng M K, Zhang W K 2003 Neural Comput. 15 349
[37]	Aubert G, Blanc-Féraud L, March R 2004 SIAM J. Numer. Anal. 42 1128
[38]	Nikolova M, Ng M K, Zhang S, Ching W K 2008 SIAM J. Imaging Sci. 1 2
[39]	Han Y, Wang W W, Feng X C 2012 Pattern Recognition 45 363
[40]	Louchet C, Moisan L 2011 SIAM J. Imaging Sci. 4 651
[41]	Shen J H, Jung Y M 2006 Appl. Math. Opt. 53 331
[42]	Xiao L, Huang Z, Wei Z 2010 EURASIP J. Adv. Signal Process. 2010 490384
[43]	Afonso M, Bioucas-Dias J, Figueiredo M 2010 IEEE Trans. Image Process. 19 2345
[44]	Chambolle A 2004 J. Math. Imaging Vis. 20 89
[45]	Wang Z https://ece.uwaterloo.ca/～ z70wang/research/ssim/[May 15 2012]
[46]	Wang Z, Bovik A C, Sheikh H R, Simoncelli E P 2004 IEEE Trans. Image Process. 13 600
[47]	Zhang H, Wang C, Zhang B, Wu F, Yan D M 2009 High-Resolution SAR Image Target Recognition (Beijing: Science Press) pp43-44 (in Chinese)[张红, 王超, 张波, 吴樊, 闫冬梅 2009 高分辨率SAR图像目标识别 (北京: 科学出版社) 第43-44页]

施引文献

[1]	Wu Y L, Wu Z S 2014 Chin. Phys. B 23 037801
[2]	Basak K, Manjunatha M, Dutta P K 2012 Med. Biol. Eng. Comput. 50 547
[3]	Ji W J, Tong C M 2013 Chin. Phys. B 22 020301
[4]	Ma C, Gu H, Su W M, Li C Z 2014 Acta Phys. Sin. 63 028403(in Chinese)[马超, 顾红, 苏卫民, 李传中 2014 63 028403]
[5]	Argenti F, Lapini A, Bianchi T, Alparone L 2013 IEEE Geosci. Remote Sens. Magaz. 1 6
[6]	Abbott J G, Thurstone F L 1979 Ultrasonic Imaging 1 303
[7]	Chen S, Zhao H C, Zhang S N, Chen Y 2013 Acta Phys. Sin. 62 218405(in Chinese)[陈思, 赵慧昌, 张淑宁, 陈勇 2013 62 218405]
[8]	Jiang Z H, Huang S X, You X B, Xiao Y G 2014 Chin. Phys. B 23 050302
[9]	Jiang Z H, Huang S X, Shi H Q, Zhang W, Wang B 2011 Acta Phys. Sin. 60 108402(in Chinese)[姜祝辉, 黄思训, 石汉青, 张伟, 王彪 2011 60 108402]
[10]	Lee J S 1980 IEEE Trans. Pattern Anal. Mach. Intell. PAMI-2 165
[11]	Frost V S, Stiles J A, Shanmugan K S, Holtzman J C 1982 IEEE Trans. Pattern Anal. Mach. Intell. PAMI-4 157
[12]	Kuan D T, Sawchuk A A, Strand T C, Chavel P 1985 IEEE Trans. Pattern Anal. Mach. Intell. PAMI-7 165
[13]	Sun Z G, Han C Z 2010 Acta Phys. Sin. 59 3210(in Chinese)[孙增国, 韩崇昭 2010 59 3210]
[14]	Donoho D L 1995 IEEE Trans. Inform. Theory 41 613
[15]	Sveinsson J R, Benediktsson J A 2003 IEEE Trans. Geosci. Remote Sens. 41 2404
[16]	Perona P, Malik J 1990 IEEE Trans. Pattern Anal. Mach. Intell. 12 629
[17]	Yu Y J, Acton S T 2002 IEEE Trans. Image Process. 11 1260
[18]	Aja-Fernández S, Alberola-López C 2006 IEEE Trans. Image Process. 15 2694
[19]	Liu G J, Zeng X P, Tian F C, Li Z Z, Chaibou K 2009 Signal Process. 89 2233
[20]	Rudin, L, Lions P L, Ohser S 2003 Geometric Level Set Methods in Imaging, Vision, and Graphics (New York: Springer) pp103-119
[21]	Aubert G, Aujol J F 2008 SIAM J Appl. Math. 68 925
[22]	Shi J, Osher S 2008 SIAM J. Imaging Sci. 1 294
[23]	Bioucas-Dias J M, Figueiredo M A T 2009 IEEE Int. Conf. Image Processing Cairo, Egypt, November 7-10 2009 p3717
[24]	Bioucas-Dias J M, Figueiredo M A T 2010 IEEE Trans. Image Process. 19 1720
[25]	Huang Y M, Ng M K, Wen Y W 2009 SIAM J. Imaging Sci. 2 20
[26]	Steidl G, Teuber T 2010 J. Math Imaging Vis. 36 168
[27]	Yun S, Woo H 2012 IEEE Trans. Image Process. 21 2523
[28]	Woo H, Yun S 2012 IEEE Trans. Image Process. 21 1701
[29]	Han Y, Feng X C, Baciu G, Wang W W 2013 Pattern Recognition 46 989
[30]	Chen D Q, Cheng L Z 2012 IEEE Trans. Image Process. 21 1650
[31]	Wang X D, Feng X C, Huo L G 2012 Acta Automat. Sin. 38 444(in Chinese)[王旭东, 冯象初, 霍雷刚 2012 自动化学报 38 444]
[32]	Chen D Q, Cheng L Z 2014 J. Comput. Appl. Math. 257 29
[33]	Li J C, Ma Z H, Peng Y X, Huang B 2013 Acta Phys. Sin. 62 099501(in Chinese)[李金才, 马自辉, 彭宇行, 黄斌 2013 62 099501]
[34]	Sun Z G 2013 Acta Phys. Sin. 62 180701(in Chinese)[孙增国 2013 62 180701]
[35]	Nikolova M, Ng M K, Tam C P 2010 IEEE Trans. Image Process. 19 3073
[36]	Kreutz-Delgado K, Murray J F, Rao B D, Engan K, Ng M K, Zhang W K 2003 Neural Comput. 15 349
[37]	Aubert G, Blanc-Féraud L, March R 2004 SIAM J. Numer. Anal. 42 1128
[38]	Nikolova M, Ng M K, Zhang S, Ching W K 2008 SIAM J. Imaging Sci. 1 2
[39]	Han Y, Wang W W, Feng X C 2012 Pattern Recognition 45 363
[40]	Louchet C, Moisan L 2011 SIAM J. Imaging Sci. 4 651
[41]	Shen J H, Jung Y M 2006 Appl. Math. Opt. 53 331
[42]	Xiao L, Huang Z, Wei Z 2010 EURASIP J. Adv. Signal Process. 2010 490384
[43]	Afonso M, Bioucas-Dias J, Figueiredo M 2010 IEEE Trans. Image Process. 19 2345
[44]	Chambolle A 2004 J. Math. Imaging Vis. 20 89
[45]	Wang Z https://ece.uwaterloo.ca/～ z70wang/research/ssim/[May 15 2012]
[46]	Wang Z, Bovik A C, Sheikh H R, Simoncelli E P 2004 IEEE Trans. Image Process. 13 600
[47]	Zhang H, Wang C, Zhang B, Wu F, Yan D M 2009 High-Resolution SAR Image Target Recognition (Beijing: Science Press) pp43-44 (in Chinese)[张红, 王超, 张波, 吴樊, 闫冬梅 2009 高分辨率SAR图像目标识别 (北京: 科学出版社) 第43-44页]

[1]	邢健崇, 张文静, 杨涛. 玻色-爱因斯坦凝聚中的非正则涡旋态及其动力学. , 2023, 72(10): 100306. doi: 10.7498/aps.72.20222289
[2]	蒋乐昕, 谢振龙, 郭泽虹, 丘伊宁, 陈溢杭. 基于准正则模式的全电介质超材料宽带反射器机理. , 2023, 72(20): 204205. doi: 10.7498/aps.72.20230915
[3]	李宁, TuXin, 黄孝龙, 翁春生. 基于Tikhonov正则化参数矩阵的激光吸收光谱燃烧场二维重建光路设计方法. , 2020, 69(22): 227801. doi: 10.7498/aps.69.20201144
[4]	周大方, 张树林, 蒋式勤. 用于心脏电活动成像的空间滤波器输出噪声抑制方法. , 2018, 67(15): 158702. doi: 10.7498/aps.67.20180294
[5]	余波, 何秋燕, 袁晓. 任意阶标度分形格分抗与非正则格型标度方程. , 2018, 67(7): 070202. doi: 10.7498/aps.67.20171671
[6]	周宏强, 万玉红, 满天龙. 基于位相变更的非相干数字全息自适应成像. , 2018, 67(4): 044202. doi: 10.7498/aps.67.20172202
[7]	张艳艳, 陈苏婷, 葛俊祥, 万发雨, 梅永, 周晓彦. 自适应非凸稀疏正则化下自适应光学系统加性噪声的去除. , 2017, 66(12): 129501. doi: 10.7498/aps.66.129501
[8]	刘波, 王青, 李永明, 隆正文. 有限空间中经典场的正则量子化. , 2015, 64(10): 100301. doi: 10.7498/aps.64.100301
[9]	王青, 隆正文, 罗翠柏. 有限空间中Dirac场的正则量子化. , 2013, 62(10): 100305. doi: 10.7498/aps.62.100305
[10]	李金才, 马自辉, 彭宇行, 黄斌. 基于图像熵的各向异性扩散相干斑噪声抑制. , 2013, 62(9): 099501. doi: 10.7498/aps.62.099501
[11]	李金才, 黄思训, 彭宇行, 张卫民. 一种用于合成孔径雷达图像相干斑抑制的双边滤波参数配置新方法. , 2012, 61(11): 119501. doi: 10.7498/aps.61.119501
[12]	何明元, 杜华栋, 龙智勇, 黄思训. 大气廓线参数反演中基于大气可反演指数的正则化参数选择方法. , 2012, 61(2): 024205. doi: 10.7498/aps.61.024205
[13]	李金才, 黄斌, 彭宇行. 一种改进的用于合成孔径雷达图像相干斑抑制的双边滤波参数配置方法. , 2012, 61(18): 189501. doi: 10.7498/aps.61.189501
[14]	盛峥, 黄思训. 变分伴随正则化方法从雷达回波反演海洋波导(Ⅰ):理论推导部分. , 2010, 59(3): 1734-1739. doi: 10.7498/aps.59.1734
[15]	盛峥, 黄思训. 变分伴随正则化方法从雷达回波反演海洋波导(Ⅱ):实际反演试验. , 2010, 59(6): 3912-3916. doi: 10.7498/aps.59.3912
[16]	谢月新, 李志坚, 周光辉. 介观耗散电容耦合电路量子化中的正则变换. , 2007, 56(12): 7224-7229. doi: 10.7498/aps.56.7224
[17]	赵仁, 张丽春, 张胜利. 正则黑洞熵与相变. , 2007, 56(12): 7355-7358. doi: 10.7498/aps.56.7355
[18]	赵仁, 张丽春, 张胜利. 正则黑洞熵. , 2007, 56(7): 3719-3722. doi: 10.7498/aps.56.3719
[19]	乔永芬, 赵淑红. 非保守系统广义Raitzin正则方程的形式不变性与非Noether守恒量. , 2006, 55(2): 499-503. doi: 10.7498/aps.55.499
[20]	乔永芬, 张耀良, 赵淑红. 完整非保守系统Raitzin正则运动方程的积分因子和守恒定理. , 2002, 51(8): 1661-1665. doi: 10.7498/aps.51.1661

计量

文章访问数: 5541
PDF下载量: 693
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

搜索

留言板