基于非局部均值滤波的SAR图像去斑
|
摘要
SAR特殊的成像机制导致图像相干斑噪声的产生,使得图像信噪比下降,为目标识别和特征提取研究造成困难,因此去除相干斑噪声的预处理是一个不可缺少的过程。本文利用非局部均值滤波算法对SAR图像去斑进行了一定的研究工作。
(1)分析非局部均值滤波的算法思想,提出了适合SAR图像乘性相干斑噪声模型的基于分类的非局部均值去斑算法。该算法采用比值算子检测算法将SAR图像像素按点、线、边缘和面进行分类,根据分类结果采用不同策略的非局部均值滤波进行去斑。同时,针对SAR图像中线与边缘的多方向特性,利用方向信息提出一种具有旋转不变性的相似度计算方法。该方法可以搜索得到更多的相似像素,有利于获得更稳健的估计。
(2)针对比值算子检测算法的不足,提出了新的检测算法。该模板具有固定窗口尺寸和统一的门限,而且具有更多的方向信息,因此能够以较小的计算代价获得更优的性能。实验证明本文算法在保证检测到的线和边缘目标连贯的同时,检测到的错误线或边缘相对于比值检测算法要少。
(3)从非局部均值算法中的相似性度量公式出发,提出了新的像素点相似性的度量方法。通过已知的观测灰度值和噪声标准差,该方法可以计算出像素邻域间的真实灰度距离。该算法克服了现有SAR图像去斑结果中细节保持与平滑程度的矛盾问题,而且该算法操作简单,易于实现。
本文工作得到了国家自然科学基金(No.6050510、60702062)的支持。
The special imaging mechanism of SAR leads speckle, which cause difficulties in the research of target recognition and feature extraction. Therefore, despeckling is an indispensable process. This dissertation studies the nonlocal means (NL-means) algorithm and applies it for SAR despeckling.
A new NL-means despeckling method for SAR image is proposed, which is adapted to the multiplicative model of speckle noise. By this method, image pixels are first classified into different classes such as point, line, edge, surface, etc., using ratio edge detector. Then, different smooth parameters of NL-means filter are used according to the class information. In addition, a searching method for rotation-invariant similar patches is designed through the use of directional information, which improves the accuracy of similar patches searching.
According to the disadvantages of ratio edge detector, a new detection template which has a fixed window size, uniform threshold, and more direction information is proposed. It can get better performance with smaller computational cost. Experiments prove that this algorithm can guarantee the coherence of edge and line, at the same time the amout of error line and edge targe is less compared with ratio edge detector.
A new similarity measure formula is deduces for SAR image by researching the similarity measure formula of NL-means filtering. Real gray distance of pixels neighbor can be calculated by the observant gray value and the noise standard deviation. This algorithm overcome the problem between details maintain and smoothing degree Moreover, the algorithm is simple and easy to realize.
The research is supported by NSFC(No.6050510,60702062).
(1)分析非局部均值滤波的算法思想,提出了适合SAR图像乘性相干斑噪声模型的基于分类的非局部均值去斑算法。该算法采用比值算子检测算法将SAR图像像素按点、线、边缘和面进行分类,根据分类结果采用不同策略的非局部均值滤波进行去斑。同时,针对SAR图像中线与边缘的多方向特性,利用方向信息提出一种具有旋转不变性的相似度计算方法。该方法可以搜索得到更多的相似像素,有利于获得更稳健的估计。
(2)针对比值算子检测算法的不足,提出了新的检测算法。该模板具有固定窗口尺寸和统一的门限,而且具有更多的方向信息,因此能够以较小的计算代价获得更优的性能。实验证明本文算法在保证检测到的线和边缘目标连贯的同时,检测到的错误线或边缘相对于比值检测算法要少。
(3)从非局部均值算法中的相似性度量公式出发,提出了新的像素点相似性的度量方法。通过已知的观测灰度值和噪声标准差,该方法可以计算出像素邻域间的真实灰度距离。该算法克服了现有SAR图像去斑结果中细节保持与平滑程度的矛盾问题,而且该算法操作简单,易于实现。
本文工作得到了国家自然科学基金(No.6050510、60702062)的支持。
The special imaging mechanism of SAR leads speckle, which cause difficulties in the research of target recognition and feature extraction. Therefore, despeckling is an indispensable process. This dissertation studies the nonlocal means (NL-means) algorithm and applies it for SAR despeckling.
A new NL-means despeckling method for SAR image is proposed, which is adapted to the multiplicative model of speckle noise. By this method, image pixels are first classified into different classes such as point, line, edge, surface, etc., using ratio edge detector. Then, different smooth parameters of NL-means filter are used according to the class information. In addition, a searching method for rotation-invariant similar patches is designed through the use of directional information, which improves the accuracy of similar patches searching.
According to the disadvantages of ratio edge detector, a new detection template which has a fixed window size, uniform threshold, and more direction information is proposed. It can get better performance with smaller computational cost. Experiments prove that this algorithm can guarantee the coherence of edge and line, at the same time the amout of error line and edge targe is less compared with ratio edge detector.
A new similarity measure formula is deduces for SAR image by researching the similarity measure formula of NL-means filtering. Real gray distance of pixels neighbor can be calculated by the observant gray value and the noise standard deviation. This algorithm overcome the problem between details maintain and smoothing degree Moreover, the algorithm is simple and easy to realize.
The research is supported by NSFC(No.6050510,60702062).
引文
[1]谢寿生,徐永进.微波遥感技术与应用.北京：科学出版社,1987.
[2]汤子跃,张守融.双站合成孔径雷达系统原理.北京：科学出版设,2003.
[3]Maitre H.合成孔径雷达图像处理.孙洪等译.北京：电子工业出版社,2005：67-69.
[4]Lee J S. Digital Image Smoothing and the Sigma Filter. Computer Vision, Graphics and Image Processing,1983,24:255-269.
[5]Kuan D.T, Sawchuk A A, Strand T C, et al. Adaptive restauration of images with speckle. IEEE Trans. ASSP,1987,35(3):373-383.
[6]Victor S F, Josephine A S. A model for radar images and its application to adaptive digital filtering of multiplicative noise. IEEE Trans. PAMI,1982,4(2):157-165.
[7]Jose M B D, Tiago A M S, and Jose M N-Leitao. Adaptive Restoration of Speckled SAR Images. IGARSS'98,1998,19-23.
[8]张微,孙蓉桦,章孝灿.基于改进的小波软阈值法的SAR图像去噪.遥感信息.2003,41(8)：1773-1784.
[9]庄镇泉,解光军,李军等.SAR图像平稳小波变换相干斑抑制方法.中国科学技术大学学报,2002,32(5)：5-572.
[10]Simoncelli E P, Adlson E H. Noise removal via Bayesian wavelet coring. IEEE Int. Conf. on Image Processing,1996,379-382.
[11]Hua X, Leland E P, and Fawwaz T U. SAR speckle reduction using wavelet denoising and Markov randomfield modeling. IEEE Trans. GARS,2002, 40(10):2196-2212.
[12]Do M N, Vetterli M. Contourlets:A new directional multiresolution image representation. Conference Record of the Thirty-Sixth Asilomar Conference on Signals, Systems and Computers,2002,1:497-501.
[13]Do M N, Directional multiresolution image representations.EPFL,2001.
[14]Velisavljevic V, Beferullo B, Vetterli M, et al. Directionlets:anisotropic multidirectional representation with separable filtering. IEEE Trans Image Process, 2006,15(7):1916-1933.
[15]Pennec Le E, Mallat S. Non linear image approximation with bandelets. CMAP Ecole Polytechnique,2003.
[16]Pennec Le E, Mallat S. Sparse geometric image representation with bandelets. IEEE Trans. Image Processing,2005,14(4):423-438.
[17]Ridgelet C E J. Theory and Applications:Ph D Thesis. Department of Statistics, Stanford University,1998.
[18]Greenspan H, et al. Rotation invariant texture recognition using a streerable pyramid. Proc.12th ICPR, Ⅱ. Jerusalem,1994,162-167.
[19]De Rivaz P F C. Complex Wavelet Based Image Analysis and Synthesis. Cambridge:University of Cambridge,2000.
[20]Lee J S. A simple speckle smoothing algorithm for synthetic aperture radar images. IEEE Trans. System, Man,and Cybernatics,1998, SMC-13:85-89.
[21]Oliver C. Understanding Synthetic Aperture Radar Images. Artech House, Boston/London,1998.
[22]Lance M. Kaplan, Analysis of Multiplicative Speckle Models for Template-Based SAR ATR, IEEE Trans. On AER.Vol.37, No.4, October,2001.
[23]Thhr M, Chin K.C, and Goodman J.W. When is speckle noise multiplicative. Appl. Opt.,1982, Vol.21,1157-1159.
[24]Mallat S. A Wavelet Tour of Signal Processing.2nd ed. New York:Academic Press,1999.
[25]Lopes A, Nezry E, Touzi R. Maximum a posteriori speckle filtering and first order texture model in SAR images. Proc. of the IGASS'90,1990,2409-2412.
[26]Lopes A, Nezry E, Touzi R. Structure detection and statistical adaptive speckle filters in SAR images. Int. J. Remote Sensing,1993,14(9):1735-1758.
[27]Lee J S. Speckle analysis and smoothing of synthetic radar images. Computer Graphics and Image Processing,1981,17:24-32.
[28]Lopes A, Touzi R, Nezry E. Adaptive speckle filters and scene heterogeneity. IEEE Trans. Geosci. Remote Sens.,1990,28(6):992-1000.
[29]Chang, Yu B, and Vetterli M. Adaptive wavelet thresholding for image denoising and compression. IEEE Trans. Image Processing, September 2000,9(9):1532-1546.
[30]Achim A, Tsakalides P, Bezerianos A. SAR image denoising via Bayesian wavelet shrinkage based on heavy-tailed modeling. IEEE Transactions on Geoscience and Remote Sensing,2003,41(8):1773-1784.
[31]Antonini M, Barlaud M, and Mathieu P. Image coding using wavelet transform. IEEE Trans. Image Processing, April 1992, 1(2):205-220.
[32]Wainwright M J, Simoncelli E P, and Willsky A S. Random cascades on wavelet trees and their use in modeling and analyzing natural imagery. Appl. Comput. Harmon. Anal., July 2001,11(1):89-123.
[33]Gleich D, Datcu M. Wavelets-besed despeckling of SAR images using Gauss-Markov random fields. IEEE Trans. Geosci. Remote Sens.,2007,45(12): 4127-4143.
[34]Donoho D. De-noising by soft-thresholding. IEEE Transactions on Information Theory,1995,41:613-627.
[35]Starck J L, Martagh F, Candes E J, et al. Gray and color image contract enhancement by the Curvelet transform. IEEE Trans Image Process,2003,12 (6):706-717.
[36]Buades, B. Coll, and J. M. Morel. A non-local algorithm for image denoising. Proc. IEEE CVPR,2005,2:60-65.
[37]Awate S A, Whitaker R T. Image denoising with unsupervised, information-theoretic, adaptive filtering. Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition,2005, to appear.
[38]Mahmoudi M, Sapiro G. Fast image and video denoising via nonlocal means of similar neighborhoods. IEEE Signal Processing Letters,2005,12 (12):839-842.
[39]Wang J, Guo Y W, and Ying Y T, et al. Fast non-local algorithm for image denoising. Proceedings of the International Conference on Image Processing, Atlanta,2006,1429-1432.
[40]Charles Kervrann, Jerome Boulanger, and Pierrick Coupe. Bayesian nonlocal means filter, image redundancy and adaptive dictionaries for noise removal, in Proceedings of the 1st International Conference on Scale Space and Variational Methods in Computer Vision (SSVM'07),2007 May-June.520-532.
[41]章毓晋.图像处理和分析.北京：清华大学出版社,1999.
[42]Touzi R, Lopes A, Bousquet P. A Statistical and Geometrical Edge Detector for SAR Image Segmentation. In:Proceeding of IGARSS,1987,1469-1474.
[43]Touzi R, Lopes A, Bousquet P. A Statistical and Geometrical Edge Detector for SAR Images. IEEE Transcation Geoscience and Remote Sensing,1988,26(6): 764-773.
[44]Argenti F, Alparone L. Speckle removal from SAR images in the undecimated wavelet domain. IEEE Trans. Geosci. Remote Sensing,2002,11(40),2363-2374.
[45]Averbuch A, Coifman R R, Donoho D L, et al. Fast Slant Stack:A notion of radon transform for data in a cartesian grid which is rapidly computible, algebraically exact, geometrically faithful an invertible. Department of Statisics, Stanford University,2001.
[46]Candes E J. Ridgelets:estimating with ridge functions. Department of Statistics Stanford University,1999.
[47]Touzi R, Lopes A, and Bousquet P. A Statistical and Geometrical Edge Detector for SAR Images. IEEE Transcation Geoscience and Remote Sensing,1988,26(6): 764-773.
[48]Lee J S, Miller A R, and Hoppel K W. Statistics of phase difference and product magnitude of multilook processed Gaussian signals. Waves Random Media, 1994.vol.4.
[49]Foucher S, Benie G B, and Boucher J M. Multiscale MAP filtering of SAR images. IEEE Trans. Image Processing,2001, vol.10,49-60.
[50]Oliver C J. A model for non-rayleigh scattering statistics.Opt. Acta,1984,31: 701-722.
[51]Dai M, Peng C, Chan A K, et al. Bayesian wavelet shrinkage with edge detection for SAR images despeckling. IEEE Trans. Geosci. Remote Sens,2004,42(8): 1642-1648.
[52]Oliver C, Quegan S. Understanding Synthetic Aperture Radar Images. Boston: ArtechHouse,1998.
[53]Lindenbaum M, Fischer M, and Bruckstein A. On gabor contribution to image enhancement. Pattern Recognition,1994,27:1-8.
[54]Alvarez L, Lions P L, and Morel J M. Image selective smoothing and edge detection by nonlinear diffusion (ⅱ).Journal of numerical analysis,1992,29: 845-866.
[55]Perona P, Malik J. Scale space and edge detection using anisotropic diffusion. IEEE Trans. Patt. Anal. Mach. Intell.1990,12:629-639.
[56]Tomasi C, Manduchi R. Bilateral filtering for gray and color images. In Proceedings of the Sixth Internatinal Conference on Computer Vision,1998, 839-846,
[57]宋宇,李满天,孙立宁.基于相似度函数的图像椒盐噪声自适应滤除算法.自动化学报.2007,5(33)：474-479.
[2]汤子跃,张守融.双站合成孔径雷达系统原理.北京：科学出版设,2003.
[3]Maitre H.合成孔径雷达图像处理.孙洪等译.北京：电子工业出版社,2005：67-69.
[4]Lee J S. Digital Image Smoothing and the Sigma Filter. Computer Vision, Graphics and Image Processing,1983,24:255-269.
[5]Kuan D.T, Sawchuk A A, Strand T C, et al. Adaptive restauration of images with speckle. IEEE Trans. ASSP,1987,35(3):373-383.
[6]Victor S F, Josephine A S. A model for radar images and its application to adaptive digital filtering of multiplicative noise. IEEE Trans. PAMI,1982,4(2):157-165.
[7]Jose M B D, Tiago A M S, and Jose M N-Leitao. Adaptive Restoration of Speckled SAR Images. IGARSS'98,1998,19-23.
[8]张微,孙蓉桦,章孝灿.基于改进的小波软阈值法的SAR图像去噪.遥感信息.2003,41(8)：1773-1784.
[9]庄镇泉,解光军,李军等.SAR图像平稳小波变换相干斑抑制方法.中国科学技术大学学报,2002,32(5)：5-572.
[10]Simoncelli E P, Adlson E H. Noise removal via Bayesian wavelet coring. IEEE Int. Conf. on Image Processing,1996,379-382.
[11]Hua X, Leland E P, and Fawwaz T U. SAR speckle reduction using wavelet denoising and Markov randomfield modeling. IEEE Trans. GARS,2002, 40(10):2196-2212.
[12]Do M N, Vetterli M. Contourlets:A new directional multiresolution image representation. Conference Record of the Thirty-Sixth Asilomar Conference on Signals, Systems and Computers,2002,1:497-501.
[13]Do M N, Directional multiresolution image representations.EPFL,2001.
[14]Velisavljevic V, Beferullo B, Vetterli M, et al. Directionlets:anisotropic multidirectional representation with separable filtering. IEEE Trans Image Process, 2006,15(7):1916-1933.
[15]Pennec Le E, Mallat S. Non linear image approximation with bandelets. CMAP Ecole Polytechnique,2003.
[16]Pennec Le E, Mallat S. Sparse geometric image representation with bandelets. IEEE Trans. Image Processing,2005,14(4):423-438.
[17]Ridgelet C E J. Theory and Applications:Ph D Thesis. Department of Statistics, Stanford University,1998.
[18]Greenspan H, et al. Rotation invariant texture recognition using a streerable pyramid. Proc.12th ICPR, Ⅱ. Jerusalem,1994,162-167.
[19]De Rivaz P F C. Complex Wavelet Based Image Analysis and Synthesis. Cambridge:University of Cambridge,2000.
[20]Lee J S. A simple speckle smoothing algorithm for synthetic aperture radar images. IEEE Trans. System, Man,and Cybernatics,1998, SMC-13:85-89.
[21]Oliver C. Understanding Synthetic Aperture Radar Images. Artech House, Boston/London,1998.
[22]Lance M. Kaplan, Analysis of Multiplicative Speckle Models for Template-Based SAR ATR, IEEE Trans. On AER.Vol.37, No.4, October,2001.
[23]Thhr M, Chin K.C, and Goodman J.W. When is speckle noise multiplicative. Appl. Opt.,1982, Vol.21,1157-1159.
[24]Mallat S. A Wavelet Tour of Signal Processing.2nd ed. New York:Academic Press,1999.
[25]Lopes A, Nezry E, Touzi R. Maximum a posteriori speckle filtering and first order texture model in SAR images. Proc. of the IGASS'90,1990,2409-2412.
[26]Lopes A, Nezry E, Touzi R. Structure detection and statistical adaptive speckle filters in SAR images. Int. J. Remote Sensing,1993,14(9):1735-1758.
[27]Lee J S. Speckle analysis and smoothing of synthetic radar images. Computer Graphics and Image Processing,1981,17:24-32.
[28]Lopes A, Touzi R, Nezry E. Adaptive speckle filters and scene heterogeneity. IEEE Trans. Geosci. Remote Sens.,1990,28(6):992-1000.
[29]Chang, Yu B, and Vetterli M. Adaptive wavelet thresholding for image denoising and compression. IEEE Trans. Image Processing, September 2000,9(9):1532-1546.
[30]Achim A, Tsakalides P, Bezerianos A. SAR image denoising via Bayesian wavelet shrinkage based on heavy-tailed modeling. IEEE Transactions on Geoscience and Remote Sensing,2003,41(8):1773-1784.
[31]Antonini M, Barlaud M, and Mathieu P. Image coding using wavelet transform. IEEE Trans. Image Processing, April 1992, 1(2):205-220.
[32]Wainwright M J, Simoncelli E P, and Willsky A S. Random cascades on wavelet trees and their use in modeling and analyzing natural imagery. Appl. Comput. Harmon. Anal., July 2001,11(1):89-123.
[33]Gleich D, Datcu M. Wavelets-besed despeckling of SAR images using Gauss-Markov random fields. IEEE Trans. Geosci. Remote Sens.,2007,45(12): 4127-4143.
[34]Donoho D. De-noising by soft-thresholding. IEEE Transactions on Information Theory,1995,41:613-627.
[35]Starck J L, Martagh F, Candes E J, et al. Gray and color image contract enhancement by the Curvelet transform. IEEE Trans Image Process,2003,12 (6):706-717.
[36]Buades, B. Coll, and J. M. Morel. A non-local algorithm for image denoising. Proc. IEEE CVPR,2005,2:60-65.
[37]Awate S A, Whitaker R T. Image denoising with unsupervised, information-theoretic, adaptive filtering. Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition,2005, to appear.
[38]Mahmoudi M, Sapiro G. Fast image and video denoising via nonlocal means of similar neighborhoods. IEEE Signal Processing Letters,2005,12 (12):839-842.
[39]Wang J, Guo Y W, and Ying Y T, et al. Fast non-local algorithm for image denoising. Proceedings of the International Conference on Image Processing, Atlanta,2006,1429-1432.
[40]Charles Kervrann, Jerome Boulanger, and Pierrick Coupe. Bayesian nonlocal means filter, image redundancy and adaptive dictionaries for noise removal, in Proceedings of the 1st International Conference on Scale Space and Variational Methods in Computer Vision (SSVM'07),2007 May-June.520-532.
[41]章毓晋.图像处理和分析.北京：清华大学出版社,1999.
[42]Touzi R, Lopes A, Bousquet P. A Statistical and Geometrical Edge Detector for SAR Image Segmentation. In:Proceeding of IGARSS,1987,1469-1474.
[43]Touzi R, Lopes A, Bousquet P. A Statistical and Geometrical Edge Detector for SAR Images. IEEE Transcation Geoscience and Remote Sensing,1988,26(6): 764-773.
[44]Argenti F, Alparone L. Speckle removal from SAR images in the undecimated wavelet domain. IEEE Trans. Geosci. Remote Sensing,2002,11(40),2363-2374.
[45]Averbuch A, Coifman R R, Donoho D L, et al. Fast Slant Stack:A notion of radon transform for data in a cartesian grid which is rapidly computible, algebraically exact, geometrically faithful an invertible. Department of Statisics, Stanford University,2001.
[46]Candes E J. Ridgelets:estimating with ridge functions. Department of Statistics Stanford University,1999.
[47]Touzi R, Lopes A, and Bousquet P. A Statistical and Geometrical Edge Detector for SAR Images. IEEE Transcation Geoscience and Remote Sensing,1988,26(6): 764-773.
[48]Lee J S, Miller A R, and Hoppel K W. Statistics of phase difference and product magnitude of multilook processed Gaussian signals. Waves Random Media, 1994.vol.4.
[49]Foucher S, Benie G B, and Boucher J M. Multiscale MAP filtering of SAR images. IEEE Trans. Image Processing,2001, vol.10,49-60.
[50]Oliver C J. A model for non-rayleigh scattering statistics.Opt. Acta,1984,31: 701-722.
[51]Dai M, Peng C, Chan A K, et al. Bayesian wavelet shrinkage with edge detection for SAR images despeckling. IEEE Trans. Geosci. Remote Sens,2004,42(8): 1642-1648.
[52]Oliver C, Quegan S. Understanding Synthetic Aperture Radar Images. Boston: ArtechHouse,1998.
[53]Lindenbaum M, Fischer M, and Bruckstein A. On gabor contribution to image enhancement. Pattern Recognition,1994,27:1-8.
[54]Alvarez L, Lions P L, and Morel J M. Image selective smoothing and edge detection by nonlinear diffusion (ⅱ).Journal of numerical analysis,1992,29: 845-866.
[55]Perona P, Malik J. Scale space and edge detection using anisotropic diffusion. IEEE Trans. Patt. Anal. Mach. Intell.1990,12:629-639.
[56]Tomasi C, Manduchi R. Bilateral filtering for gray and color images. In Proceedings of the Sixth Internatinal Conference on Computer Vision,1998, 839-846,
[57]宋宇,李满天,孙立宁.基于相似度函数的图像椒盐噪声自适应滤除算法.自动化学报.2007,5(33)：474-479.