摘要
合成孔径雷达(SAR)具有全天候、全天时以及穿透能力强等特点,在军事和民用领域均有着广泛的应用。然而,SAR成像信息不仅与地物类型有关,也与雷达观测角、地物复介电常数、雷达发射波的波长、极化方式等因素有关,使得单个波段单种极化方式下的SAR系统获取的信息非常有限,随着相关硬件技术的突破,SAR系统可以获取多个波段和极化方式下对同一场景的观测信息,多源SAR图像信息的融合应用得到越来越多的关注。地物分类和目标检测是SAR图像解译的两个典型应用,在“十一五”预研项目的支持下,本文重点研究了如何利用多源SAR图像信息来提高地物分类率和目标检测精度,主要内容如下:
1.高分辨率全极化SAR图像背景杂波建模与参数估计,SAR图像杂波背景建模是SAR图像处理分析的基础,在目前已有的极化SAR统计模型中, G p0分布能够较好的对复杂场景建模,然而G p0分布参数估计并没有得到很好的解决,针对这一问题,本文提出了基于对数累积量的参数估计方法,对实测数据的拟合实验表明,该方法无论在拟合速度和拟合精度上相比已有的最大似然估计和矩估计方法都有所提高,改进的杂波分布将用在后续的SAR图像地物分类和目标检测中。
2.多波段SAR地物分类与决策融合,不同波段数据具有不同的极化方式,需要使用不同的分类算法,对单极化数据的分类,本文使用应用广泛的SVM方法,对于全极化数据的分类,在对最具代表性的Lee方法深入分析的基础上提出了一种改进的全极化SAR图像分类算法,由于该方法引入了螺旋散射和更好的杂波模型,因此在一些复杂场景中取得了比Lee方法更好的分类效果。利用改进的算法实现了不同波段数据的分类,实验结果表明不同波段数据在区分不同地物类型上存在一定的互补性,因此在Dempster-Shafer证据理论的框架下,进一步实现了不同波段数据的分类决策级融合,结果证实多个波段信息的融合可以获得比单个波段更好的分类效果。在对分类结果进行评价时,因为真实的专题制图信息难以获取,本文以容易获取的Google earth影像为参考,由于使用了整个场景的样本数据,所以比传统的人工选择测试样本的方法更客观,更准确。
3.伪装网遮蔽车辆检测与决策融合,本文在课题的支持下,在国内率先开展了伪装网遮蔽车辆检测试验,对不同波段极化方式下的遮蔽车辆散射功率进行了深入对比研究,发现入射波越长,目标散射功率越强,在同一波段的不同极化中,交叉极化下的目标成像效果好于同极化。恒虚警检测算法(CFAR)是目前最有可能实用的一类方法,在该算法中,背景杂波统计模型是决定算法性能的关键因素,结合前面改进的杂波模型对全局CFAR算法做了改进,利用改进算法分别实现了有伪装网和无伪装网时不同波段和极化方式下的车辆检测,并对检测结果进行基于Neyman-Pearson准则的决策级融合。实验表明伪装网能够起到一定的遮蔽效果,降低目标检测率,不同波段和极化检测结果的决策级融合可以提高目标检测率,降低虚警率,提升综合检测性能,而且有网时的检测性能提升要高于无网,这意味着伪装网情况下进行多波段多极化SAR检测融合更有必要。
4.机场环境解译及算法界面介绍,在前面地物分类和目标检测的研究基础上对机场环境这一典型军事场景进行了处理,得到机场环境的初步解译结果,有利于进一步战场分析和态势估计。最后,给出了上述算法在平台中的相关界面实现。
The Synthetic Aperture Radar (SAR) has been widely used in military and civilian areas, because of its weather-independent, all time and strong penetrating capability. However, the radar backscattering information is not only related to terrain types, but also influenced by other factors such as radar observation angle, terrain complex dielectric constant, incident wavelength, polarization etc. These factors lead to the limited capability of obtaining information for single-band single-polarization SAR. With the breakthrough of hardware technology, SAR system can image under multi-bands and multi-polarization modes simultaneously. Therefore, the study of multichannel SAR image fusion has attracted more and more attention. Terrain classification and target detection are two typical applications of SAR image interpretation. Under the support of "Eleventh Five-Year" National Defense pre-research project, this paper focuses on how to use multiple sources SAR images to improve the terrain classification and target detection accuracy. The main work and contribution are as follows:
1. The statistical model of background clutter and parameters estimation is studied, which forms the basis of SAR image analysis. Among the existing polarimetric SAR models, G p0 is a good model for characterizing polarimetric SAR images containing different terrain types. Nevertheless, parameters estimation hasn't been well resolved. To solve this problem, this paper proposed a new method based on Log-det cumulant (LDC), experiments on real SAR data demonstrate its superiority over the maximum likehood method (ML) and moment based method (MoM) in terms of speed and accuracy. This laid the foundation for SAR image analysis such as classification and target detection.
2. Multiband SAR terrain classification and decision level fusion are studied. Different bands data have different polarization properties, requiring different classfication methods. For single polarimetric SAR data, the classical SVM method is adopted. For fully polarimetric SAR data, by introducing more suitable scattering model and better clutter distribution, we proposed an improved algorithm for polarimetric SAR image classification and obtained better results than Lee's. Moreover, the experiment results indicate that different band data are complementary in distinguishing terrain, so fusion of the classification results at desicion level is realized under the Dempster-Shafer evidence theory framework. Experiments show that the integration of multiple bands information could get better classification results than single band. To evaluate the performance of classification algorithm, cooresponding Google earth image is used as reference instead of real thematic mapping information which is difficult to obtain. Due to the use of entire scene sample data, this method appears more objective than the traditional method of artificial selection testing samples.
3. Under the support of National Defence project, this paper takes the lead in camouflage covered vehicle detection. After indepth study of scattering power by covered vehicles under different bands and polarization, we found that the echo power is stronger by longer incident wavelength and more visible by co-polarization wave than cross-polarization wave. Constant False Alarm Rate (CFAR) target detection algorithm is one of the most potential methods. For this algorithm, the background clutter modelling is a key factor in influencing the algorithm performance. A modified global CFAR algorithm is proposed by adopting the improved clutter model and is applied to veiled and unveiled vehicles detection under different band and polarization modes. Detection results are fusioned at decision level according to Neyman- Pearson criterion, which show that the camouflage net could reduces the target detection rate to some extent. Fusion of different band and polarization detection results at decision level could improve the accuracy, reduce false alarm rate and raise the overall detection performance. Moreover, detection performance is raised higher for camouflage covered vehicles than exposed vehicles, which means that multi-band multi-polarization detection fusion is more meaningful for veiled target detection.
4. Based on the above proposed classification and target detection algorithms, a typical military scene (airport) is processed and the preliminary interpretation results are obtained, which benefits further battlefield surveillance and situation assessment. Finally, these algorithm interfaces in our software platform are presented.
引文
[1]卢丹,洪峻.合成孔径雷达多波段的应用及效果评价[J],测绘科技动态,1998,20(2):24-29
[2]李春升,李景文,周荫清.空载合成孔径雷达技术及展望[J],电子学报,1995,23(10):156-159
[3]焦李成,王爽,候彪. SAR图像理解和解译研究进展[J],电子学报,2005,33(12):2423-2434
[4] D.M.Mckeown, W.A.Harvey, J.McDermott. Rule-based interpretation of aerial imagery[J]. IEEE Trans.on PAMI, 1985,7(5):570-585
[5] V.Clement,G.Giraudon, S.Houzelle. Interpretation of remotely sensed images in a context of multisensor fusion using a multispecialist architecture[J]. IEEE Trans.on GeoSci&Remote Sensing, 1993,31(7):779-791
[6] T.Matsuyama. Knowledge-based aerial image understanding systems and expert systems for image processing[C].In Procedings of IGARSS’05,Zurich,1986:1026- 1038
[7] M.R.Toma, F.Vinelli. Processing of polarimetric and multifrequency sar data recorded by maestro-1campaign[J]. IEEE Trans.on GeoSci&Remote Sensing, 1991,3(7):79-91
[8] M.Craig Dobson,Leland E.Pierce. Knowledge-based land-cover classification using ERS-1/JERS-1 SAR composites, IEEE Trans.on Geosci &Remote Sensing, 1996,34(1): 83-99
[9] F.Sery,D.Ducrot-Gambart. Multisource classification of SAR images with the use of segmentation, polarimetric, texture and multitemporal data[C]. proc.SPIE Europto, Taormina,Italy,1996:186-197
[10] Sylvie Le Hegarat Mascle,Issabelle Bloch,and D.Vidal.Madjar. Application of Dempster-Shafer Evidence Theory to Unsupervised Classification in Multisource Remote Sensinig[J], IEEE Trans.on Geosci&RemoteSensiing, 1997,35(4): 1018-1031
[11] W.Hagg, M.Sties, Monitoring the Oder/Germany flood with ERS, RADARSAT and optical data[C], Proceedings of IGRSS’98, 1998:1614-1615
[12] E.Mougin, C.Proisy,G.Marty etc, Multifrequency and Multipolarization Radar Backscattering from Mangrove Forests[J], IEEE Trans.on GeoSci&Remote Sensing, 1999, 37(1):94-102
[13] F.Tupin,I.Bloch,H.Maitre. A First step toward automatic interpretation of sar images using evidential fusion of several structure detectors[J]. IEEE trans.on GeoSci&Remote Sensing, 1999,37(3):1327-1343
[14] Ferro-Famil L.,Pottier E.and Lee J.S. Unsupervised classification of multifrequency and fully polarimetric SAR images based on the H/A Alpha-Wishart classifier[J], IEEE Trans on Geoscience and Remote Sensing,2001,39(11):2332-2342
[15]苏芳,洪文,毛士艺.贝叶斯融合在SAR图像分类中的应用[J],电子学报,2003,31(7):1107-1110
[16] Kouskoulas Y,Ulaby FT,Pierce L E. The Bayesian hierarchical classifier and its application to short vegetation using muhifrequency polarimetric SAR[J]. IEEE TGRS 2004,469-477
[17] S.Paloscia,G.Macelloni, The contribution of Multitemporal SAR data in assessing Hydrological parameters[J],IEEE trans on Geoscience and Remote Sensing, 2004,1(3):201-205
[18] D.Borghys, Y.Yvinec,C.Perneel. Supervised Feature-based classification of multi-channel sar images[J],Pattern Recognition Letters,2006
[19] Alejandro C.Frery, Antonio H.Correia, Corina da C.Freitas, Classifying Multifrequency Fully Polarimetric Imagery With Multiple Sources of Statistical Evidence and Contextual Information[J], IEEE Trans on Geoscience and Remote Sensing, 2007,45(10):3098-3109
[20]刘向君,多波段SAR目标检测与图像分类融合[D],长沙:国防科技大学,2005
[21] Bjorn Waske, Matthias Braun, Classifier ensembles for land cover mapping using multitemporal SAR imagery[J], ISPRS Journal of Photogrammetry and Remote Sensing, 2009,64(1): 450-457
[22] C.C.Freitas,A.C.Frery,and A.H.Correia.The polarimetric G distribution for SAR data analysis[J],Environmetrics,2005,16(1):13-31.
[23] A C Frery, H Muller, C F Yanasse. A model for extremely heterogeneous clutter[J]. IEEE Trans. on GRS, 1997,35(3):648-659.
[24] J.S.Lee, E.Pottier, Polarimetric radar imaging: from basics to application[J], Boca Raton, FL:CRC press, 2009
[25] B Epstein. Some applications of the Mellin transform in statistics[J], The Annals of Mathematical Statistics, 1948,19(3):370-379
[26] I.S. Gradshteyn, I.M.Ryzhik, Table of Integrals, Series, and Products[M]. 6thed.San Diego, CA: Academic, 2000
[27] S. N. Anfinsen, A. P. Doulgeris. Estimation of the equivalent number of looks in PolSAR imagery[C], in Proc. IGARSS'08, Boston, USA, 2008, 487-490.
[28] MD DeVore, O’Sullivan JA. Quantitative statistical assessment of conditional models for synthetic aperture radar[J].IEEE Trans. Image Process, 2004,13(2):113-125
[29]赵书河,多源遥感影像决策级融合及其应用研究[D],南京:南京大学,2003
[30] Lein J.K. Applying Evidential Reasoning Methods to Agricultural Land Cover Classification[J]. Int.J.Remote Sensing, 2003,24(21):4161-4180
[31] Min.Sil.Yang, Wooil M.Moon, Decision Level Fusion of MultiFrequecy polarimetric SAR and Optical data with Dempster-Shafer Evidence Theory[C], in proceedings of GRSS, 2003,3668-3670
[32]胡德勇,李京,陈云浩等,单波段单极化SAR图像水体和居民地信息提取方法研究[J],中国图像图形学报,2008,13(2):257-263
[33]李海权,李春霞等,基于纹理和BP神经网络的SAR图像分类[J],遥感信息,2009,3(2):124-128
[34] Robert M. Haralick, Statistical and structural approaches to texture[C], in Proceedings of the IEEE,1979, 67(5):786-804
[35] Vapnik V N.Statistical Learning Theory[M].New York:John Viley&Sons Inc,1998
[36] Q Zhao,J S Principe, V Brennan etc.Synthetic aperture radar automatic target recognition with three strategies of learning and representation[J], Optical Engineering, 2000, 39:1230-1244
[37] M Bryant, F Garber. SVM classifier applied to the MSTAR public data set[C]. In Proc.SPIE:Algorithms for Synthetic Aperture Radar Imagery VI.Orlando Florida,1999:355-360
[38] Su FuLin, NiLiang,Li Da Fang,etal. Classification of SAR image based on gray coocurrence matrix and support vector machine[C].In Proceedings of 7th International Conference on Signal Processing,Harbin,China,2004,2:1385-1388
[39] AngiulliG, Barrile V, Cacciola M. SAR imagery classification using multi-class support vector machines[J]. Journal of Electromagetic Waves and Applications, 2005, 19(14):1865-1872
[40] Hsu Chih-Wei, Lin CHih-Jen. A comparison of methods for multi-class support vector mathines[J]. IEEE trans. On Neural Networks,2002,13(2):415-425
[41] J.J.Van Zyl, Unsupervised classification of scattering mechanisms using radar polarimetric data[J], IEEE Trans.Geosci.RemoteSens, 1989, 27(1): 36-45.
[42] J.S.Lee, M.R.Grunes, and R.Kwok, Classification of multi-look polarimetric SAR imagery based on the complex Wishart distribution[J], Int.J.RemoteSens., 1994,15(11): 2299-2311.
[43] LeeJ.S,Grunes M.R, Pottier E etc. Unsupervised terrain classification preserving polarimetric scattering characteristics[J], IEEE Trans.Geosci.Remote Sensing,2004, 42(4):722-731
[44] LeeJ.S,Grunes M.R.Aniworth T.L, Unsupervised classification using polarimetric decomposition and the complex Wishart classifier[J],IEEE Trans.on Geosci and Remote Sensing,1999,37(5):2249-2258
[45] Pottier E, Lee J.Ferro-family. Advanced concepts in polarimetry-part2 : polarimetric target classification[M]. In:Radar Polarimetry and Interferometry. Educational Notes RTO-EN-SET-081bits,France,2007
[46]郑志彬,叶中付,基于相位相关的图像配准算法[J],数据采集与处理,2006,21(4):444-449
[47] Lee,JS.Digital image enhancement and noise filtering by use of local statistics[J],IEEE Transon Pattern Analysis and Machine Intelligence,1980(2):165-168
[48] Yuki Yajima,Yoshio Yamaguchi,Ryoichi Sato,etc. PolSAR image analysis of wetlands using a modified four-component scattering power decomposition[J], IEEE Trans on Geosci and Remote sensing, 2008,46(6):1667-1673
[49] http://www.csie.ntu.edu.tw/~cjlin/libsvm/
[50] Duarte,M.F., Yu Hen Hu, Optimal decision fusion with applications to target detection in wireless ad hoc sensor networks[C], 6th workshop on Multimedia signal processing, 2004,319-322
[51]周春花.基于DS证据推理的目标舰船融合检测研究[D],西安:西北工业大学,2004
[52]匡纲要,高贵,蒋咏梅等.合成孔径雷达目标检测理论、算法及应用[M],长沙:国防科技大学出版社,2007
[53]高贵,鲁敏,黄纪军等.高分辨SAR图像中杂波统计特性分析[J],信号处理,2008,24(4):63-69
[54]贺志国,周晓光,陆军,匡纲要.一种基于G0分布的SAR图像快速CFAR检测方法[J],国防科技大学学报,2009,31(1):47-51