基于形态成分分析的复杂背景SAR图像舰船尾迹检测
|
摘要
SAR图像舰船尾迹检测不仅可用于反演运动舰船的航速航向信息,也有助于发现远小于尾迹的弱小舰船目标.针对现有基于直线检测的舰船尾迹检测方法一般仅适用于简单海况背景SAR图像的问题,提出一种基于形态成分分析的复杂背景SAR图像舰船尾迹检测方法.该方法依据各形态成分只在特定字典下稀疏表示的基本原则,针对舰船尾迹的"线"奇异性及海面纹理的"点"奇异性分别选择不同的基函数构建字典,实现了稀疏意义下舰船尾迹结构成分与海面纹理成分的分离;通过剪切波高频系数重构,实现了舰船尾迹结构成分的增强;最终,使用Radon变换进行舰船尾迹线检测与定位.定性与定量实验结果表明,复杂海况背景下,采用该方法进行舰船尾迹检测的效果明显优于现有其他方法.
The detection of ship wakes in SAR images is helpful not only for estimating the speed and the direction of moving ships, but also for finding ship objects which are far smaller than ship wakes. Facing the problem that the existing detection methods based on the line detection can be only used for SAR images under the simple sea background conditions, a novel ship wake detection method for SAR images with the complex backgrounds using the morphological component analysis is proposed. Relied on the basic principle that each component can only be sparsely represented by a specific dictionary, suitable basis functions are selected for the dictionary construction according to the "line" singularity of ship wakes and the "point" singularity of sea-surface textures respectively, and the ship wake structure component and the sea-surface texture component are consequently separated in the sense of sparse representation. Then, the enhancement of ship wake structure component is realized through the reconstruction of the high-frequency coefficients of the shearlets transform. And finally, the ship wake lines are detected by using the Radon transform. Both the qualitative and the quantitative experimental results show that, the proposed method conspicuously outperforms the existing methods for the ship wake detection in SAR images, especially under the complex sea background conditions.
The detection of ship wakes in SAR images is helpful not only for estimating the speed and the direction of moving ships, but also for finding ship objects which are far smaller than ship wakes. Facing the problem that the existing detection methods based on the line detection can be only used for SAR images under the simple sea background conditions, a novel ship wake detection method for SAR images with the complex backgrounds using the morphological component analysis is proposed. Relied on the basic principle that each component can only be sparsely represented by a specific dictionary, suitable basis functions are selected for the dictionary construction according to the "line" singularity of ship wakes and the "point" singularity of sea-surface textures respectively, and the ship wake structure component and the sea-surface texture component are consequently separated in the sense of sparse representation. Then, the enhancement of ship wake structure component is realized through the reconstruction of the high-frequency coefficients of the shearlets transform. And finally, the ship wake lines are detected by using the Radon transform. Both the qualitative and the quantitative experimental results show that, the proposed method conspicuously outperforms the existing methods for the ship wake detection in SAR images, especially under the complex sea background conditions.
引文
[1]Lyden J D,Hammond R R,Lyzenga D R,et al.Synthetic aperture radar imaging of surface ship wakes[J].Journal of Geophysical Research:Oceans,1988,93(C10):12293-12303
[2]Rey M T,Tunaley J K,Folinsbee J T,et al.Application of radon transform techniques to wake detection in seasat-A SAR images[J].IEEE Transactions on Geoscience and Remote Sensing,1990,28(4):553-560
[3]Chong Jinsong,Zhu Minhui.Survey of the study on ship and wake detection in SAR imagery[J].Acta Electronica Sinica,2003,31(9):1356-1360(in Chinese)(种劲松,朱敏慧.SAR图像舰船及其尾迹检测研究综述[J].电子学报,2003,31(9):1356-1360)
[4]Kuo J M,Chen K S.The application of wavelets correlator for ship wake detection in SAR images[J].IEEE Transactions on Geoscience and Remote Sensing,2003,41(6):1506-1511
[5]Courmontagne P.An improvement of ship wake detection based on the BBRadon transform[J].Signal Processing,2005,85(8):1634-1654
[6]Chong Jinsong,Zhu Minhui.Ship wake detection algorithm in SAR image based on normalized grey level Hough transform[J].Journal of Image and Graphics,2004,9(2):146-150(in Chinese)(种劲松,朱敏慧.基于归一化灰度Hough变换SAR图像舰船尾迹检测算法[J].中国图象图形学报,2004,9(2):146-150)
[7]Ai J Q,Qi X Y,Yu W D,et al.A novel ship wake CFAR detection algorithm based on SCR enhancement and normalized hough transform[J].IEEE Geoscience and Remote Sensing Letters,2011,8(4):681-685
[8]Xing X W,Ji K F,Zou H X,et al.An enhancing normalized Radon transform method for ship wake detection in SAR imagery[C]//Proceedings of the 9th European Conference on Synthetic Aperture Radar.Frankfurt:VDE Press,2012:559-562
[9]Zhang Mingzhao,Meng Tao,Mu Jianhua,et al.Method for detecting ship trail of ocean synthetic aperture radar image:China,CN102542277A[P].2012.07.04(in Chinese)(张明照,孟涛,牟建华,等.一种海洋合成孔径雷达图像的舰船尾迹检测方法:中国,CN102542277A[P].2012.07.04)
[10]Copeland A C,Ravichandran G,Trivedi M M.Localized radon transform-based detection of ship wakes in SAR images[J].IEEE Transactions on Geoscience and Remote Sensing,1995,33(1):35-45
[11]Wang Shiqing,Jin Yaqiu.Ship wake detection in SAR images based on Radon transformation and morphologic image processing[J].Journal of Remote Sensing,2001,5(4):289-294(in Chinese)(王世庆,金亚秋.SAR图像船行尾迹检测的Radon变换和形态学图像处理技术[J].遥感学报,2001,5(4):289-294)
[12]Mata-Moya D,Jarabo-Amores P,Jimenez-Chaparro B,et al.Application of mean-shift filtering to ship wakes detection in SAR images[C]//Proceedings of the 8th European Conference on Synthetic Aperture Radar.Frankfurt:VDE Press,2010:1-4
[13]Tang Ziyue,Zhu Minhui,Wang Weiyan.A CFAR detection method of ship wakes in SAR images[J].Acta Electronica Sinica,2002,30(9):1336-1339(in Chinese)(汤子跃,朱敏慧,王卫延.一种SAR图象舰船尾迹的CFAR检测方法[J].电子学报,2002,30(9):1336-1339)
[14]Graziano M D.SAR-based ship route estimation by wake components detection and classification[C]//Proceedings of IEEE International Geoscience and Remote Sensing Symposium.Los Alamitos:IEEE Computer Society Press,2015:3255-3258
[15]Nan J,Wang C,Zhang B,et al.Ship wake CFAR detection algorithm in SAR images based on length normalized scan[C]//Proceedings of IEEE International Geoscience and Remote Sensing Symposium.Los Alamitos:IEEE Computer Society Press,2013:3562-3565
[16]Fadili J M,Starck J L,Elad M,et al.MCALab:reproducible research in signal and image decomposition and inpainting[J].Computing in Science&Engineering,2010,12(1):44-63
[17]He Z,Bystrom M.The chordlet transform with an application to shape compression[J].Signal Processing:Image Communication,2012,27(2):140-152
[18]Sun J,Ren G Q,Wu Q Z.The easy block Hadamard transform:a new adaptive directional Hadamard transform for sparse image representation[J].International Journal for Light and Electron Optics,2014,125(10):2356-2360
[19]Aharon M,Elad M,Bruckstein A.K-SVD:an algorithm for designing overcomplete dictionaries for sparse representation[J].IEEE Transactions on Signal Processing,2006,54(11):4311-4322
[20]Eksioglu E M,Bayir O.K-SVD meets transform learning:transform K-SVD[J].IEEE Signal Processing Letters,2014,21(3):347-351
[21]Eksioglu E M.Online dictionary learning algorithm with periodic updates and its application to image denoising[J].Expert Systems with Applications,2014,41(8):3682-3690
[22]Wang L Z,Lu K,Liu P,et al.IK-SVD:dictionary learning for spatial big data via incremental atom update[J].Computing in Science Engineering,2014,16(4):41-52
[23]Pan Zongxu,Yu Jing,Xiao Chuangbai,et al.Dictionary learning and structural self-similarity-based codebook mapping for single image super resolution[J].Journal of Computer-Aided Design&Computer Graphics,2015,27(6):1032-1038(in Chinese)(潘宗序,禹晶,肖创柏,等.基于字典学习与结构自相似性的码本映射超分辨率算法[J].计算机辅助设计与图形学学报,2015,27(6):1032-1038)
[24]Starck J L,Elad M,Donoho D L.Image decomposition via the combination of sparse representations and a variational approach[J].IEEE Transactions on Image Processing,2005,14(10):1570-1582
[25]Kutyniok G,Labate D.Introduction to shearlets[M]//Shearlets:Multiscale Analysis for Multivariate Data.Berlin:Birkh?user Press,2012:1-38
[26]Easley G,Labate D,Lim W Q.Sparse directional image representations using the discrete shearlet transform[J].Applied and Computational Harmonic Analysis,2008,25(1):25-46
[27]Agarwal S,Awan A,Roth D.Learning to detect objects in images via a sparse,part-based representation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(11):1475-1490
[2]Rey M T,Tunaley J K,Folinsbee J T,et al.Application of radon transform techniques to wake detection in seasat-A SAR images[J].IEEE Transactions on Geoscience and Remote Sensing,1990,28(4):553-560
[3]Chong Jinsong,Zhu Minhui.Survey of the study on ship and wake detection in SAR imagery[J].Acta Electronica Sinica,2003,31(9):1356-1360(in Chinese)(种劲松,朱敏慧.SAR图像舰船及其尾迹检测研究综述[J].电子学报,2003,31(9):1356-1360)
[4]Kuo J M,Chen K S.The application of wavelets correlator for ship wake detection in SAR images[J].IEEE Transactions on Geoscience and Remote Sensing,2003,41(6):1506-1511
[5]Courmontagne P.An improvement of ship wake detection based on the BBRadon transform[J].Signal Processing,2005,85(8):1634-1654
[6]Chong Jinsong,Zhu Minhui.Ship wake detection algorithm in SAR image based on normalized grey level Hough transform[J].Journal of Image and Graphics,2004,9(2):146-150(in Chinese)(种劲松,朱敏慧.基于归一化灰度Hough变换SAR图像舰船尾迹检测算法[J].中国图象图形学报,2004,9(2):146-150)
[7]Ai J Q,Qi X Y,Yu W D,et al.A novel ship wake CFAR detection algorithm based on SCR enhancement and normalized hough transform[J].IEEE Geoscience and Remote Sensing Letters,2011,8(4):681-685
[8]Xing X W,Ji K F,Zou H X,et al.An enhancing normalized Radon transform method for ship wake detection in SAR imagery[C]//Proceedings of the 9th European Conference on Synthetic Aperture Radar.Frankfurt:VDE Press,2012:559-562
[9]Zhang Mingzhao,Meng Tao,Mu Jianhua,et al.Method for detecting ship trail of ocean synthetic aperture radar image:China,CN102542277A[P].2012.07.04(in Chinese)(张明照,孟涛,牟建华,等.一种海洋合成孔径雷达图像的舰船尾迹检测方法:中国,CN102542277A[P].2012.07.04)
[10]Copeland A C,Ravichandran G,Trivedi M M.Localized radon transform-based detection of ship wakes in SAR images[J].IEEE Transactions on Geoscience and Remote Sensing,1995,33(1):35-45
[11]Wang Shiqing,Jin Yaqiu.Ship wake detection in SAR images based on Radon transformation and morphologic image processing[J].Journal of Remote Sensing,2001,5(4):289-294(in Chinese)(王世庆,金亚秋.SAR图像船行尾迹检测的Radon变换和形态学图像处理技术[J].遥感学报,2001,5(4):289-294)
[12]Mata-Moya D,Jarabo-Amores P,Jimenez-Chaparro B,et al.Application of mean-shift filtering to ship wakes detection in SAR images[C]//Proceedings of the 8th European Conference on Synthetic Aperture Radar.Frankfurt:VDE Press,2010:1-4
[13]Tang Ziyue,Zhu Minhui,Wang Weiyan.A CFAR detection method of ship wakes in SAR images[J].Acta Electronica Sinica,2002,30(9):1336-1339(in Chinese)(汤子跃,朱敏慧,王卫延.一种SAR图象舰船尾迹的CFAR检测方法[J].电子学报,2002,30(9):1336-1339)
[14]Graziano M D.SAR-based ship route estimation by wake components detection and classification[C]//Proceedings of IEEE International Geoscience and Remote Sensing Symposium.Los Alamitos:IEEE Computer Society Press,2015:3255-3258
[15]Nan J,Wang C,Zhang B,et al.Ship wake CFAR detection algorithm in SAR images based on length normalized scan[C]//Proceedings of IEEE International Geoscience and Remote Sensing Symposium.Los Alamitos:IEEE Computer Society Press,2013:3562-3565
[16]Fadili J M,Starck J L,Elad M,et al.MCALab:reproducible research in signal and image decomposition and inpainting[J].Computing in Science&Engineering,2010,12(1):44-63
[17]He Z,Bystrom M.The chordlet transform with an application to shape compression[J].Signal Processing:Image Communication,2012,27(2):140-152
[18]Sun J,Ren G Q,Wu Q Z.The easy block Hadamard transform:a new adaptive directional Hadamard transform for sparse image representation[J].International Journal for Light and Electron Optics,2014,125(10):2356-2360
[19]Aharon M,Elad M,Bruckstein A.K-SVD:an algorithm for designing overcomplete dictionaries for sparse representation[J].IEEE Transactions on Signal Processing,2006,54(11):4311-4322
[20]Eksioglu E M,Bayir O.K-SVD meets transform learning:transform K-SVD[J].IEEE Signal Processing Letters,2014,21(3):347-351
[21]Eksioglu E M.Online dictionary learning algorithm with periodic updates and its application to image denoising[J].Expert Systems with Applications,2014,41(8):3682-3690
[22]Wang L Z,Lu K,Liu P,et al.IK-SVD:dictionary learning for spatial big data via incremental atom update[J].Computing in Science Engineering,2014,16(4):41-52
[23]Pan Zongxu,Yu Jing,Xiao Chuangbai,et al.Dictionary learning and structural self-similarity-based codebook mapping for single image super resolution[J].Journal of Computer-Aided Design&Computer Graphics,2015,27(6):1032-1038(in Chinese)(潘宗序,禹晶,肖创柏,等.基于字典学习与结构自相似性的码本映射超分辨率算法[J].计算机辅助设计与图形学学报,2015,27(6):1032-1038)
[24]Starck J L,Elad M,Donoho D L.Image decomposition via the combination of sparse representations and a variational approach[J].IEEE Transactions on Image Processing,2005,14(10):1570-1582
[25]Kutyniok G,Labate D.Introduction to shearlets[M]//Shearlets:Multiscale Analysis for Multivariate Data.Berlin:Birkh?user Press,2012:1-38
[26]Easley G,Labate D,Lim W Q.Sparse directional image representations using the discrete shearlet transform[J].Applied and Computational Harmonic Analysis,2008,25(1):25-46
[27]Agarwal S,Awan A,Roth D.Learning to detect objects in images via a sparse,part-based representation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(11):1475-1490