基于特征的SAR图像自动配准方法研究
|
摘要
遥感图像的配准是将不同时相、不同遥感平台等获取的数据配准到同一坐标系统下的过程。由于各传感器成像的分辨率不同、时相不同或成像机制不同等原因,使得图像之间必然出现相对平移、旋转、比例缩放等变化,不能直接进行图像融合,必须进行图像配准,以便各个图像能进行像元与像元间的对比和运算。
合成孔径雷达具有全天时,全天候的工作特性,并且可以穿透树木,得到高分辨率的合成孔径雷达图像(SAR图像)。高分辨率的SAR图像不仅在军用方面得到了很大的发展,其民用用途也越来也广泛。这就促使了有关SAR图像解译方面的研究越来越多,也越来越深入。SAR图像的配准就是逐渐发展而来的一种重要的SAR图像处理过程,它对于SAR图像和光学图像融合,SAR图像间的变化检测等是必不可少的一个步骤。
本文在这个背景下进行了基于特征的SAR图像配准方法的研究。首先在第二章介绍了有关遥感图像配准的理论知识,然后第三章重点介绍了SAR图像配准的研究现状和配准方法,并对这些方法进行分析比较,总结出具有代表性的SAR图像配准研究方法及他们的适用性,重点研究了基于特征的配准方法,同时结合了基于灰度统计特性的方法。灰度统计特性主要研究了互信息准则和对齐度准则的方法,实验得出他们的适用性。最后,在基于这些研究的基础上,提出了一种基于特征的由粗到精(coarse-to-fine)的SAR图像自动配准方法,对分割后的SAR图像寻找匹配区域并进行修正。实验证明了此方法的有效性,无论从计算时间还是从计算精度上都有一定的改善。
Remote Sensing image registration is to register data from different time or different sensor platform to one pixel system. Due to different resolution, time and imaging mechanism for different sensors, remote sensing image is inevitably moving, rotating and zooming which lead to unable fusion directly. All of these decide that image registration is necessary. Synthetic Aperture Radar (SAR) can work all day, all-weather, and can go through the trees to get high resolution SAR image. High resolution SAR image is not only using widely in army application, but also in civil application. More and more research on SAR image is deep.SAR image registration is a necessary step of SAR image processing. It is essential for the fusion between SAR image and optical image, and SAR image detection and so on.
This thesis does work based on these states. Firstly, chapter 2 introduces the principle knowledge about remote sensing image registration. And then, chapter 3 has a detailed description on the present condition and methods of SAR image registration. After comparing these algorithms and analyzing them, some conclusions can be obtained. At the same time, experiments on Mutual Information and Alignment Metric methods show their application in SAR image registration. Based on these theories, chapter 4 proposes a coarse-to-fine feature-based method for SAR image registration. The segmentation process and the match process have considered the speckle noise of SAR image. The experiment shows it applicable and effective, no matter from the time consuming and precision. At last, it concludes the above content and gives an expectation for SAR image registration.
合成孔径雷达具有全天时,全天候的工作特性,并且可以穿透树木,得到高分辨率的合成孔径雷达图像(SAR图像)。高分辨率的SAR图像不仅在军用方面得到了很大的发展,其民用用途也越来也广泛。这就促使了有关SAR图像解译方面的研究越来越多,也越来越深入。SAR图像的配准就是逐渐发展而来的一种重要的SAR图像处理过程,它对于SAR图像和光学图像融合,SAR图像间的变化检测等是必不可少的一个步骤。
本文在这个背景下进行了基于特征的SAR图像配准方法的研究。首先在第二章介绍了有关遥感图像配准的理论知识,然后第三章重点介绍了SAR图像配准的研究现状和配准方法,并对这些方法进行分析比较,总结出具有代表性的SAR图像配准研究方法及他们的适用性,重点研究了基于特征的配准方法,同时结合了基于灰度统计特性的方法。灰度统计特性主要研究了互信息准则和对齐度准则的方法,实验得出他们的适用性。最后,在基于这些研究的基础上,提出了一种基于特征的由粗到精(coarse-to-fine)的SAR图像自动配准方法,对分割后的SAR图像寻找匹配区域并进行修正。实验证明了此方法的有效性,无论从计算时间还是从计算精度上都有一定的改善。
Remote Sensing image registration is to register data from different time or different sensor platform to one pixel system. Due to different resolution, time and imaging mechanism for different sensors, remote sensing image is inevitably moving, rotating and zooming which lead to unable fusion directly. All of these decide that image registration is necessary. Synthetic Aperture Radar (SAR) can work all day, all-weather, and can go through the trees to get high resolution SAR image. High resolution SAR image is not only using widely in army application, but also in civil application. More and more research on SAR image is deep.SAR image registration is a necessary step of SAR image processing. It is essential for the fusion between SAR image and optical image, and SAR image detection and so on.
This thesis does work based on these states. Firstly, chapter 2 introduces the principle knowledge about remote sensing image registration. And then, chapter 3 has a detailed description on the present condition and methods of SAR image registration. After comparing these algorithms and analyzing them, some conclusions can be obtained. At the same time, experiments on Mutual Information and Alignment Metric methods show their application in SAR image registration. Based on these theories, chapter 4 proposes a coarse-to-fine feature-based method for SAR image registration. The segmentation process and the match process have considered the speckle noise of SAR image. The experiment shows it applicable and effective, no matter from the time consuming and precision. At last, it concludes the above content and gives an expectation for SAR image registration.
引文
[1]程芳,朱敏慧,吴一戎,改进的多项式图像配准方法[J],电子与信息学报,vol.23,No.11,Sept,2001,PP:1071-1076.
[2]J.Duchon,Interpolation des fonctions de deux variables suivant le principle de la flexion des plaques minces[J],RAIRO Analytical Numericals,10,1976,pp:5-12.
[3]W.K.Pratt,Digital Image Processing[J],2nd,Wiley,New York,1991.
[4]D.I.Barnea,H,F.Silverman,A class of algorithms for fast digital image registrations[J].IEEE Transactions on Computing 21,1972,pp:179-186.
[5]P.Thevenaz,M.Unser,A pyramid approach to sub-pixel image fusion based on mutual information[J],Proceedings of the IEEE International Conference on Image Processing ICIP'96,Lausanne,Switzerland,1996,pp:265-268.
[6]P.Thevenaz,M.Unser,Spline pyramids for inter-modal image registration using mutual information[J],Proceedings of SPIE:Wavelet Applications in Signal and ImageProcessing,San Diego,CA,1997,pp:236-247.
[7]王东峰,邹谋炎.基于最大化对齐度的多模态图像自动配准[J].电子与信息学报,2003,25(6):755-76.
[8]Ranade,S,A.Rosenfeld.Point pattern matching by relaxation[J].PatternRecognition,Vol.12,No.4,1980,pp:269-275,1980.
[9]王小平,曹立明著.遗传算法--理论、应用与软件实现[M].西安:西安交通大学出版社,2002.
[10]Maes F,Vandermeulen D,and Suetens P.Comparative evaluation of multiresolution optimization strategies for multimodality image registration by maximization of mutual information[j].Medical Image Analysis,Vol.3,No.4,1999,pp:373-386.
[11]Kirkpatriek S,Gelatt C D,Veeehi M E Optimization by simulated annealing[J].Science,Vol.220,No.4598,1983,pp:671-680.
[12]王靖,朱梦宇,赵保军,何佩琨,基于小波和改进型Hausdorff距离的遥感图像配准方法[J].电子学报,vol.34,No.12,Dec.2006,pp:2167-2169.
[13]Fernin E,Nasr,J.-M,Edge Detection and Hough Transform Techniques Applied to High-relief SAR Imagery For Automatic Registration[J].IEEE,May,1990,pp:643-646.
[14]康欣,韩崇昭,杨艺,基于结构特征的SAR图像配准,系统仿真学报,vol.18,No.5,May,2006,pp:1307-1310.
[15]Jeremy S.De.Bonet,Alan Chao,Structure-driven SAR image registration[J],SPIE,Vol.3370,April,1998,pp:109-119.
[16]Wen Yang,Chuanzhao Han,Hong Sun,Yongfeng Cao,Registration of high resolution SAR and Optical Images Based on Multiple Features[J].IEEE,July, 2005,PP:3542-3544.
[17]邓鹏,边缘与灰度信息结合的SAR图像配准算法,北京:中国科学院电子学研究所工学硕士论文,2003.
[18]Xiangyu,Hong Sun,Yongfeng Cao,SAR image registration combing watershed segmentation and mutual information[J].SPIE,Vol.6403,Oct,2005,pp:60432U1-60432U8.
[19]邓鹏,王宏琦,结合边缘与灰度信息的SAR图像配准算法[J],遥感技术与应用,vol.18,No.3,Jun,2003,DD:160-163.
[20]Hiep Quang Luong,Gautama.S.Philips.W.Automatic Registration of Synthetic Aperture Radar(SAR) Images[J].IEEE,Vol.6,Sept,2004,pp:3864-3867.
[21]Wu Yifeng,MAITRE Henri,A Multi-resolution approach for registration of a spot image and a SAR image[J].IEEE,May,1990,pp:635-638.
[22]shoji Takcuchi,Image Registration Between SAR and TM Data Using DEM and SlantRange Information.Vol.3,Aug,1993,pp:1351-1353.
[23]Eugenio Sansosti,Geometrical SAR Image Registration[J].IEEE Transactions on Geoscience and Remote Sensing,Vol.44,NO.10,Oct,2006,pp:2861-2870.
[24]Bovik A.C.,On Detecting Edges in Speckle Imagery[J].IEEE Trans.on Acoustics,Speech and Signal Processing,ASSP-36(10),Oct,1988,pp:1618-1627.
[25]TOUZI R.,LOPES A.,BOUSQUET P.,A Statistical and Geometrical Edge Detector for SAR Images[J].IEEE Transactions on Geoscience and Remote Sensing,Vol.26,No.6,Nov,1988,pp:764-773.
[26]Oliver C J,Blacknell D,White R G.Optimum edge detection in SAR.IEE Proceedings on Radar,Sonar and Navigation[J],Vol.143,No.1,1996,pp:31-40.
[27]FjOrtoft R,Marthon P,Lopés A,Cubero-Castan E.An optimum multi-edge detector for SAR image segmentation[J].IEEE Transactions on Geoscience and Remote Sensing,1998,36,pp:793-802.
[28]Davidson,G.,Ouchi,K.,Saito,G.,Ishitsuka,N.,Mohri,K.,Uratsuka,S..Performance eval uation of maximum likelihood SAR segmentation for multi-temporal rice crop mapping[J].IEEE,Oct.2002,pp:390-394.
[29]Quelle,H.-C.,Delignon,Y.,Marzouki,A.Unsupervised Bayesian segmentation of SAR images using the Pearson system distributions[J].IEEE,Vol.3,Aug,1993,pp:1538-1540.
[30]Yunhan Dong,Forster,B.,Milne,A.Segmentation of radar imagery using Gaussian M arkov random field models and wavelet transform techniques[J].IEEE,Vol.4,Aug,1997,pp:2054-2056.
[31]P.Marthon,B.Paci,E.Cubero-Castan,Finding the structure of a satellite image,Proc.Europe.Image and Signal Processing for Remote Sensing[J],SPIE 2315,Rome,Italy,1994,pp:669-679.
[32]A.Lopes,R.Fjortoft,D.Ducrot,Edge detection and segmentation of SAR images in homogeneous regions[J],in Information Processing for Remote Sensing,C.H.Chen,World Scientific,pp:1-28
[33]R.Touzi,A.Lopes,P.Bousquet,A statistical and geometrical edge detector for SAR images[J].IEEE transactions on Geoscience and Remote Sensing,Vol.26,No.6,1988,pp:764-773.
[34]Gonzalez,R.C.,Woods,R.E.,Digital Image Processing,Vol.2,2002,Prentice Hall,Upper Saddle River,NJ.
[35]David Small,Francesco Holecz,Erich Meier,Daniel Nüesch,and Arnold Barmettler Proc.of Geomatics in the Era of RADARSET,Ottawa,Canada,May,1997,pp:24-30.
[36]M.K.Hu.Visual Pattern Recognition by Moment Invariants[J].IEEE Trans.Inf.Theory IT- 8,1962,pp:179-187.
[37]M.R.Teague.Image Analysis via the General Theory of Moments[J].Opt.Soc.Amer,Vol.70,1980,191a:920-930.
[38]冷雪飞,刘建业,熊智,邢广华,加权Hausdorm巨离算法在SAR/INS景象匹配中的应用[J],控制与决策,vol.21,No.1,2006,pp:42-45.
[39]Yajun Li.Reforming the Theory of Invariant Moments for Pattern Recognition[J].Pattern Recognition,Vol.25,No.7,1992,pp:723-730.
[40]Jan Flusser,On the Independence of Rotation Moment invariants"[J].Pattern Recognition.Vol.33,2000,pp:1405-1410.
[41]FLUSSER J,SUK T.A moment-based approach to registratration of Images with affine geometric distortion[J].IEEE Trans.Geosicence and Remote Sensing.Vol.32,No.2,1996,pp:382 -387.
[42]Smith S M.,Brady J M.SUSAN-A new approach to low level image processing[J].International Journal of Computer Vision,Vol.23,No.1,1997,pp:45-78.
[43]Jung K,Lacroix S.A robust interest points matching algorithmiC].Proceedings of Eighth IEEE International Conference on Computer Vision,2001,pp:538-543.
[44]Chang Yulin,Zhou Zhirnin,Chang Wenge,Jin Tian,A New Registration Method for Multi-spectral SAR Images[J].IEEE,July,2005,pp:1704-1708.
[45]Wen Yang,Chuanzhao Han,Hong Sun,Yongfeng Cao,Registration of High Resolution SAR and Optical Images Based on Multiple Features[J].IEEE,July,2005,pp:3542-3544.
[46]陈涛,图像仿射不变特征提取方法研究,长沙:国防科技大学博士学位论文,2006.
[2]J.Duchon,Interpolation des fonctions de deux variables suivant le principle de la flexion des plaques minces[J],RAIRO Analytical Numericals,10,1976,pp:5-12.
[3]W.K.Pratt,Digital Image Processing[J],2nd,Wiley,New York,1991.
[4]D.I.Barnea,H,F.Silverman,A class of algorithms for fast digital image registrations[J].IEEE Transactions on Computing 21,1972,pp:179-186.
[5]P.Thevenaz,M.Unser,A pyramid approach to sub-pixel image fusion based on mutual information[J],Proceedings of the IEEE International Conference on Image Processing ICIP'96,Lausanne,Switzerland,1996,pp:265-268.
[6]P.Thevenaz,M.Unser,Spline pyramids for inter-modal image registration using mutual information[J],Proceedings of SPIE:Wavelet Applications in Signal and ImageProcessing,San Diego,CA,1997,pp:236-247.
[7]王东峰,邹谋炎.基于最大化对齐度的多模态图像自动配准[J].电子与信息学报,2003,25(6):755-76.
[8]Ranade,S,A.Rosenfeld.Point pattern matching by relaxation[J].PatternRecognition,Vol.12,No.4,1980,pp:269-275,1980.
[9]王小平,曹立明著.遗传算法--理论、应用与软件实现[M].西安:西安交通大学出版社,2002.
[10]Maes F,Vandermeulen D,and Suetens P.Comparative evaluation of multiresolution optimization strategies for multimodality image registration by maximization of mutual information[j].Medical Image Analysis,Vol.3,No.4,1999,pp:373-386.
[11]Kirkpatriek S,Gelatt C D,Veeehi M E Optimization by simulated annealing[J].Science,Vol.220,No.4598,1983,pp:671-680.
[12]王靖,朱梦宇,赵保军,何佩琨,基于小波和改进型Hausdorff距离的遥感图像配准方法[J].电子学报,vol.34,No.12,Dec.2006,pp:2167-2169.
[13]Fernin E,Nasr,J.-M,Edge Detection and Hough Transform Techniques Applied to High-relief SAR Imagery For Automatic Registration[J].IEEE,May,1990,pp:643-646.
[14]康欣,韩崇昭,杨艺,基于结构特征的SAR图像配准,系统仿真学报,vol.18,No.5,May,2006,pp:1307-1310.
[15]Jeremy S.De.Bonet,Alan Chao,Structure-driven SAR image registration[J],SPIE,Vol.3370,April,1998,pp:109-119.
[16]Wen Yang,Chuanzhao Han,Hong Sun,Yongfeng Cao,Registration of high resolution SAR and Optical Images Based on Multiple Features[J].IEEE,July, 2005,PP:3542-3544.
[17]邓鹏,边缘与灰度信息结合的SAR图像配准算法,北京:中国科学院电子学研究所工学硕士论文,2003.
[18]Xiangyu,Hong Sun,Yongfeng Cao,SAR image registration combing watershed segmentation and mutual information[J].SPIE,Vol.6403,Oct,2005,pp:60432U1-60432U8.
[19]邓鹏,王宏琦,结合边缘与灰度信息的SAR图像配准算法[J],遥感技术与应用,vol.18,No.3,Jun,2003,DD:160-163.
[20]Hiep Quang Luong,Gautama.S.Philips.W.Automatic Registration of Synthetic Aperture Radar(SAR) Images[J].IEEE,Vol.6,Sept,2004,pp:3864-3867.
[21]Wu Yifeng,MAITRE Henri,A Multi-resolution approach for registration of a spot image and a SAR image[J].IEEE,May,1990,pp:635-638.
[22]shoji Takcuchi,Image Registration Between SAR and TM Data Using DEM and SlantRange Information.Vol.3,Aug,1993,pp:1351-1353.
[23]Eugenio Sansosti,Geometrical SAR Image Registration[J].IEEE Transactions on Geoscience and Remote Sensing,Vol.44,NO.10,Oct,2006,pp:2861-2870.
[24]Bovik A.C.,On Detecting Edges in Speckle Imagery[J].IEEE Trans.on Acoustics,Speech and Signal Processing,ASSP-36(10),Oct,1988,pp:1618-1627.
[25]TOUZI R.,LOPES A.,BOUSQUET P.,A Statistical and Geometrical Edge Detector for SAR Images[J].IEEE Transactions on Geoscience and Remote Sensing,Vol.26,No.6,Nov,1988,pp:764-773.
[26]Oliver C J,Blacknell D,White R G.Optimum edge detection in SAR.IEE Proceedings on Radar,Sonar and Navigation[J],Vol.143,No.1,1996,pp:31-40.
[27]FjOrtoft R,Marthon P,Lopés A,Cubero-Castan E.An optimum multi-edge detector for SAR image segmentation[J].IEEE Transactions on Geoscience and Remote Sensing,1998,36,pp:793-802.
[28]Davidson,G.,Ouchi,K.,Saito,G.,Ishitsuka,N.,Mohri,K.,Uratsuka,S..Performance eval uation of maximum likelihood SAR segmentation for multi-temporal rice crop mapping[J].IEEE,Oct.2002,pp:390-394.
[29]Quelle,H.-C.,Delignon,Y.,Marzouki,A.Unsupervised Bayesian segmentation of SAR images using the Pearson system distributions[J].IEEE,Vol.3,Aug,1993,pp:1538-1540.
[30]Yunhan Dong,Forster,B.,Milne,A.Segmentation of radar imagery using Gaussian M arkov random field models and wavelet transform techniques[J].IEEE,Vol.4,Aug,1997,pp:2054-2056.
[31]P.Marthon,B.Paci,E.Cubero-Castan,Finding the structure of a satellite image,Proc.Europe.Image and Signal Processing for Remote Sensing[J],SPIE 2315,Rome,Italy,1994,pp:669-679.
[32]A.Lopes,R.Fjortoft,D.Ducrot,Edge detection and segmentation of SAR images in homogeneous regions[J],in Information Processing for Remote Sensing,C.H.Chen,World Scientific,pp:1-28
[33]R.Touzi,A.Lopes,P.Bousquet,A statistical and geometrical edge detector for SAR images[J].IEEE transactions on Geoscience and Remote Sensing,Vol.26,No.6,1988,pp:764-773.
[34]Gonzalez,R.C.,Woods,R.E.,Digital Image Processing,Vol.2,2002,Prentice Hall,Upper Saddle River,NJ.
[35]David Small,Francesco Holecz,Erich Meier,Daniel Nüesch,and Arnold Barmettler Proc.of Geomatics in the Era of RADARSET,Ottawa,Canada,May,1997,pp:24-30.
[36]M.K.Hu.Visual Pattern Recognition by Moment Invariants[J].IEEE Trans.Inf.Theory IT- 8,1962,pp:179-187.
[37]M.R.Teague.Image Analysis via the General Theory of Moments[J].Opt.Soc.Amer,Vol.70,1980,191a:920-930.
[38]冷雪飞,刘建业,熊智,邢广华,加权Hausdorm巨离算法在SAR/INS景象匹配中的应用[J],控制与决策,vol.21,No.1,2006,pp:42-45.
[39]Yajun Li.Reforming the Theory of Invariant Moments for Pattern Recognition[J].Pattern Recognition,Vol.25,No.7,1992,pp:723-730.
[40]Jan Flusser,On the Independence of Rotation Moment invariants"[J].Pattern Recognition.Vol.33,2000,pp:1405-1410.
[41]FLUSSER J,SUK T.A moment-based approach to registratration of Images with affine geometric distortion[J].IEEE Trans.Geosicence and Remote Sensing.Vol.32,No.2,1996,pp:382 -387.
[42]Smith S M.,Brady J M.SUSAN-A new approach to low level image processing[J].International Journal of Computer Vision,Vol.23,No.1,1997,pp:45-78.
[43]Jung K,Lacroix S.A robust interest points matching algorithmiC].Proceedings of Eighth IEEE International Conference on Computer Vision,2001,pp:538-543.
[44]Chang Yulin,Zhou Zhirnin,Chang Wenge,Jin Tian,A New Registration Method for Multi-spectral SAR Images[J].IEEE,July,2005,pp:1704-1708.
[45]Wen Yang,Chuanzhao Han,Hong Sun,Yongfeng Cao,Registration of High Resolution SAR and Optical Images Based on Multiple Features[J].IEEE,July,2005,pp:3542-3544.
[46]陈涛,图像仿射不变特征提取方法研究,长沙:国防科技大学博士学位论文,2006.