基于物理模型的地雷检测方法研究
|
摘要
通常情况下,由于地雷目标较小且所处环境复杂,基于能量的简单检测方法虚警率过高。本文根据实际雷达系统几何模型和地雷物理特性,分析了地雷散射模型并将其参数化,得到地雷的有效特征并将其用于鉴别。
本文针对国防科学技术大学研制的车载前视超宽带地表穿透雷达,首先分析其等效单站模型及全孔径和各子孔径成像特点;然后对地雷电磁散射特性进行建模,说明双峰特征的形成机理及其方位一致性。
预筛选是目标检测重要环节,能够有效降低运算量并获取疑似目标区域,是特征提取的必要准备。本文采用常规的检测流程,包括耦合信号抑制、图像增益均衡、恒虚警检测、形态学滤波和加权聚类等。
在地雷鉴别过程中,本文选取四个与成像几何及地雷结构相关的特征,分别是方位视角偏离度,双峰特征模型中的双峰间距、双峰峰值比以及主峰宽度。根据特征确定性的特点,分别采用简单门限比较和超球面支持向量机一类分类器进行目标鉴别,在较低虚警率的情况下得到了较高的检测率。
通过实测数据验证,基于物理模型提取的特征能够有效抑制杂波,并获得较高检测率。同时,由于成像分辨率受到带宽和孔径长度影响,而其对于目标特征提取密切相关,本文又研究了APES超分辨成像和压缩感知稀疏成像以及基于超分辨图像的特征提取和鉴别方法,这也为地雷目标鉴别研究探索了一条新途径。
Generally, landmine detection method based on energy has a high false alarm rate because of the complex environment and small size of landmine itself. Based on geometric model of real radar system and physic characteristic of landmine, scattering model of landmine is analyzed and established, at the same time, efficient landmine features are extracted for discrimination.
For the real system, Stepped-frequency Forward-looking Ground-penetrating Virtual Aperture Radar developed by the National University of Defense Technology, the equal single station model together with the characteristics of sub-aperture image and full aperture image are analyzed firstly. Following that, a model of landmine electromagnetic scattering is established, and then the mechanism of double-hump characteristics together with its aspect consistency are introduced detailedly.
Filtering in advance is an important step in target detection and is essential to characteristic extraction, which can efficiently reduce operation size and help to get region of interest (ROI). In this paper, conventional flow of detection is used for extracting the target’s ROI, which include coupling signal depression, noise depression, image gain proportion, constant false alarm rate (CFAR) detection, morphological filter and weighted clustering.
In the process of landmine discrimination, four characteristics related to the system and landmine structure are adopted, including angle of view offset degree, distance between the double-hump, double-hump peaks value ratio and width of main peak in double-hump characteristic model. According to the characteristics which are decided by the physics model, target discriminators are adopted easy threshold and Hyper Sphere Support Vector Machine respectively.
Features extracted from the physics-model can be used for reducing clutter efficiently and get higher detection rate, and that is proved to be valid by the real data. Because the image revolution is limited by the bandwidth of signal and length of aperture, which makes features extraction difficultly. So new features extraction from super-revolution images using APES and Congress Sensing are studied, which is a new try in landmine detection.
本文针对国防科学技术大学研制的车载前视超宽带地表穿透雷达,首先分析其等效单站模型及全孔径和各子孔径成像特点;然后对地雷电磁散射特性进行建模,说明双峰特征的形成机理及其方位一致性。
预筛选是目标检测重要环节,能够有效降低运算量并获取疑似目标区域,是特征提取的必要准备。本文采用常规的检测流程,包括耦合信号抑制、图像增益均衡、恒虚警检测、形态学滤波和加权聚类等。
在地雷鉴别过程中,本文选取四个与成像几何及地雷结构相关的特征,分别是方位视角偏离度,双峰特征模型中的双峰间距、双峰峰值比以及主峰宽度。根据特征确定性的特点,分别采用简单门限比较和超球面支持向量机一类分类器进行目标鉴别,在较低虚警率的情况下得到了较高的检测率。
通过实测数据验证,基于物理模型提取的特征能够有效抑制杂波,并获得较高检测率。同时,由于成像分辨率受到带宽和孔径长度影响,而其对于目标特征提取密切相关,本文又研究了APES超分辨成像和压缩感知稀疏成像以及基于超分辨图像的特征提取和鉴别方法,这也为地雷目标鉴别研究探索了一条新途径。
Generally, landmine detection method based on energy has a high false alarm rate because of the complex environment and small size of landmine itself. Based on geometric model of real radar system and physic characteristic of landmine, scattering model of landmine is analyzed and established, at the same time, efficient landmine features are extracted for discrimination.
For the real system, Stepped-frequency Forward-looking Ground-penetrating Virtual Aperture Radar developed by the National University of Defense Technology, the equal single station model together with the characteristics of sub-aperture image and full aperture image are analyzed firstly. Following that, a model of landmine electromagnetic scattering is established, and then the mechanism of double-hump characteristics together with its aspect consistency are introduced detailedly.
Filtering in advance is an important step in target detection and is essential to characteristic extraction, which can efficiently reduce operation size and help to get region of interest (ROI). In this paper, conventional flow of detection is used for extracting the target’s ROI, which include coupling signal depression, noise depression, image gain proportion, constant false alarm rate (CFAR) detection, morphological filter and weighted clustering.
In the process of landmine discrimination, four characteristics related to the system and landmine structure are adopted, including angle of view offset degree, distance between the double-hump, double-hump peaks value ratio and width of main peak in double-hump characteristic model. According to the characteristics which are decided by the physics model, target discriminators are adopted easy threshold and Hyper Sphere Support Vector Machine respectively.
Features extracted from the physics-model can be used for reducing clutter efficiently and get higher detection rate, and that is proved to be valid by the real data. Because the image revolution is limited by the bandwidth of signal and length of aperture, which makes features extraction difficultly. So new features extraction from super-revolution images using APES and Congress Sensing are studied, which is a new try in landmine detection.
引文
[1] Schreiner K. Landmine detection research pushes forward, despite challenges [J]. IEEE Intelligent Systems, 2002, 17(2):4-7
[2] Maki K H. Humanitarian demining: reality and the challenge of technology–the state of the arts [J]. International Journal of Advanced Robotic Systems, 2007,4(2):151-172
[3] David J D. A review of GPR for landmine detection [J]. Sensing and Imaging: An international journal, 2006, 7(3): 90-123
[4]朱晨,李常生,孟咸勇.外军单兵探雷器发展分析[C].探雷工程侦察专业学组第六届学术年会,黑龙江满洲里, 2007, 103-108
[5] Carin, L., Geng, N., McClure, M., Sichina, J. and Nguyen, L., Ultra-wide-band synthetic-aperture radar for mine-field detection [N] IEEE Antennas and Propagation Magazine, 1999, 41(1):18-33
[6] Kaplan, L. M., Oh, S.-M. and McClellan, J. H., Multiresolution target discrimination [J] during image formation, Proceedings of SPIE, 2000, 4053:239-250
[7] Andrew J. Hill, Graeme Crisp, Justin Ratcliffe, The development of a motion-compensated, vehicle mounted, ultrawideband radar for buried landmine detection [J], Proc. of the 3rd European Radar Conf., Manchester UK, 265-268, Sept. 2006
[8] Marshall Bradley, Thomas Witten, Robert McCummins, Michael Duncan, Mine detection with ground penetrating synthetic aperture radar [J], Proc. of SPIE, 4742: 248-258, 2002
[9] Erik M. Rosen, Frank S. Rotondo, Elizabeth Ayers, Testing and evaluation of forward-looking GPR countermine systems [J], Proc. of SPIE, Bellingham, WA, 5794: 901-911,2005
[10] Novak, L. M., Halversen, S. D., Owirka, G. J. and Hiett, M., Effects of polarization and resolution on SAR ATR [J], IEEE Transactions on Aerospace and Electronic Systems, 1997, 33(1):102-116
[11]方学立, UWB-SAR图像中的目标检测与鉴别, [D],长沙:国防科学技术大学, 2005, 4
[12]金添,超宽带SAR浅埋目标成像与检测的理论和技术研究, [D],长沙:国防科学技术大学, 2007, 3
[13] Wang T P, Ozy S, and James M K, et al. Feature Analysis for Forward-Looking Landmine Detection Using GPR [J]. Proc. of SPIE , 2005, 5794: 1233-1244.
[14] Geng, N. and Carin, L., Resonances of a dielectric BOR buried in a lossy,dispersive layered medium [C], International Geoscience and Remote Sensing Symposium (IGARSS), Seattle, WA, USA, 1998:222-224
[15] Casey, K. F. and Oetzel, G. N., Physical-optics models for scatterng from bured mines [C], Second Australian-American Joint Conference on the Technology of Mine Countermeasures, Sydney, Australia, 2001:1-15
[16] Andrieu J, Gallais F, and Mallepeyre V, et al. Land mine detection with an ultra-wideband SAR system [J]. Proceedings of SPIE, 2002, 4742:237-247.
[17] Sun Y J and Li J. Landmine detection using forward-looking ground penetrating radar [J]. Proceedings of SPIE, 2005, 5794: 1089-1097.
[18] Sun Y J and Li J. Adaptive learning approach to landmine detection [J]. IEEE Transactions on Aerospace and Electronic Systems, 2005, 41(3): 973-985.
[19] Allen, M. R., Efficient approach to physics-based FOPEN-SAR ATD/R [J], Proceedings of SPIE, 1996, 2757:163-172
[20] Chaney, R. D., Willsky, A. S. and Novak, L. M., Coherent aspect-dependent SAR image formation [J], Proceedings of SPIE, 1994, 2230:256-274
[21] Allen, M. R., Jauregui, J. M. and Hoff, L. E., FOPEN-SAR detection by direct use of simple scattering physics [C], IEEE National Radar Conference, Alexandria, VA, USA, 1995:152-157
[22]蒋咏梅,叶簇穿透超宽带SAR目标检测方法研究, [D],长沙:国防科学技术大学, 1998, 10.
[23] Koklnozi A, Hosgood B, and Sieber A J. Landmine detection using a time-sequence of thermal infrared images [J]. Proceedings of SPIE, 2003, 4885:170-179
[24] Cheng J and Miller E. Model-based principal component techniques for detection of buried landmines in multiframe synthetic aperture radar images. 2002:334-336
[25]王正明等, SAR图像提高分辨率技术[M],科学出版社,2006
[26]王鹏宇,超宽带机载步进频率SAR成像方法研究, [D],长沙:国防科学技术大学,2008, 11
[27]杨延光,周智敏,宋千.分裂孔径发射阵列接收BP算法方位分辨率分析[J].信号处理, 2008, 24(5):752-756
[28]王建,车载前视超宽带SAR浅埋目标成像技术研究, [D],长沙:国防科学技术大学, 2008, 10
[29]刘光平,超宽带合成孔径雷达高效成像算法, [D],长沙:国防科学技术大学, 2003, 03
[30]朱国富,董臻,梁甸农.超宽带LFM信号的BP成像算法[J].信号处理, 2001,17(5): 424-428
[31]聂在平等,目标与环境电磁散射特性建模[M],国防工业出版社, 2009
[32]文舸一.计算电磁学的进展与展望[J].电子学报, 1995,23(10):62-69
[33] K S Yee. Numerical solution of initial boundary value problems involving Maxwell’s equations in isotropic media [J]. IEEE Trans, Ap-14(4): 302-306,1966
[34]葛德彪等,电磁波时域有限差分方法[M],西安:西安电子科技大学出版社,2002
[35] Harrington R.F. Time-Harmonic Electromagnetic Fields. New York [DB]: McGraw-Hill, 1968
[36]孙晓坤,埋地地雷超宽带电磁散射特性及在目标检测中的应用, [D],长沙:国防科学技术大学, 2008, 7
[37] Kell R E. On the Derivation of Bistatic RCS from Monostatic Measurements [J]. Proc. of the IEEE, 1965(8): 983~988
[38] Sun Y J and Li J. Time-frequency analysis for plastic landmine detection via forward-looking ground penetrating radar [J]. IEE Proc.-Radar Sonar Navig, August 2003, 150(4): 253-261.
[39]杨延光,基于车载FLGPSAR序列图像的浅埋目标检测技术研究, [D],长沙:国防科学技术大学, 2008, 11
[40] Sinan B, John G. Unsupervised iterative detection of landmines in highly cluttered environments [J]. IEEE Transaction on Image Progressing, 2003, 12(5): 509-523
[41]王玉明,车载前视超宽带GPSAR杂波特性与目标特征研究, [D],长沙:国防科学技术大学,2008, 11
[42]杨志国,基于ROI的UWB SAR叶簇覆盖目标鉴别方法研究, [D],长沙:国防科学技术大学, 2007, 9
[43]谢冰,贺志毅.雷达导引头真实地杂波数据分析[J].制导与引信,2007, 28(1): 7-10,24
[44]王小鹏等译,形态学图像分析原理与应用[M],清华大学出版社, 2008
[45] Vitebskiy S and Carin L Short-pulse scattering from buried wires and bodies of revolution [C]. Proceedings of SPIE, 2002, 2747:233-244
[46] Jin Tian, Zhou Zhi-min, Song Qian, Chang Wen-ge. A novel SVM for ground penetrating synthetic aperture radar landmine detection [J]. Journal Of Electronics (CHINA), 2008, 25(1): 70-75
[47] Li J. and Stoical P. Efficient An adaptive filtering approach to spectral estimation and SAR imaging [J]. IEEE Trans on signal processing, Vol.44, No.6, 1996, pp. 1469:1484
[48] Hongbin Li, Jian Li, Petre Stoica, Performance analysis of forward-backward matched-filterbank spectral estimators [J], IEEE Transactions on Signal Processing, 1998, 46(7):1954-1966
[49]任璐,关于提高SAR图像质量的研究, [D],西安电子科技大学,2005, 1
[50]金添,宋千,孙晓坤,周智敏.地表穿透合成孔径雷达浅地表金属地雷二维电磁特征研究[J],电子学报, 2006, 34(12):2246-2249
[51] Donoho D L. Compressed sensing [J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289?1306
[52] Candes E, Tao T, Decoding by linear programming [J], IEEE Transactions on Information Theory, 2005, 51(12):4203-4215
[53] Chen S B, Donoho D L, Saunders M A. Atomic decomposition by basis pursuit [J]. SIAM Journal on Scientific Computing, 1998, 20(1): 33-61
[54]黄琼,屈乐乐,吴秉横,方广有.压缩感知在超宽带雷达成像中的应用.电波科学学报[J], 2010, 25(1):77-80
[55] Candes E and Tao T. The dantzig selector: statistical estimation when p is much larger than n [J]. The Annals of Statistics, 2007, 35(6): 1-41
[2] Maki K H. Humanitarian demining: reality and the challenge of technology–the state of the arts [J]. International Journal of Advanced Robotic Systems, 2007,4(2):151-172
[3] David J D. A review of GPR for landmine detection [J]. Sensing and Imaging: An international journal, 2006, 7(3): 90-123
[4]朱晨,李常生,孟咸勇.外军单兵探雷器发展分析[C].探雷工程侦察专业学组第六届学术年会,黑龙江满洲里, 2007, 103-108
[5] Carin, L., Geng, N., McClure, M., Sichina, J. and Nguyen, L., Ultra-wide-band synthetic-aperture radar for mine-field detection [N] IEEE Antennas and Propagation Magazine, 1999, 41(1):18-33
[6] Kaplan, L. M., Oh, S.-M. and McClellan, J. H., Multiresolution target discrimination [J] during image formation, Proceedings of SPIE, 2000, 4053:239-250
[7] Andrew J. Hill, Graeme Crisp, Justin Ratcliffe, The development of a motion-compensated, vehicle mounted, ultrawideband radar for buried landmine detection [J], Proc. of the 3rd European Radar Conf., Manchester UK, 265-268, Sept. 2006
[8] Marshall Bradley, Thomas Witten, Robert McCummins, Michael Duncan, Mine detection with ground penetrating synthetic aperture radar [J], Proc. of SPIE, 4742: 248-258, 2002
[9] Erik M. Rosen, Frank S. Rotondo, Elizabeth Ayers, Testing and evaluation of forward-looking GPR countermine systems [J], Proc. of SPIE, Bellingham, WA, 5794: 901-911,2005
[10] Novak, L. M., Halversen, S. D., Owirka, G. J. and Hiett, M., Effects of polarization and resolution on SAR ATR [J], IEEE Transactions on Aerospace and Electronic Systems, 1997, 33(1):102-116
[11]方学立, UWB-SAR图像中的目标检测与鉴别, [D],长沙:国防科学技术大学, 2005, 4
[12]金添,超宽带SAR浅埋目标成像与检测的理论和技术研究, [D],长沙:国防科学技术大学, 2007, 3
[13] Wang T P, Ozy S, and James M K, et al. Feature Analysis for Forward-Looking Landmine Detection Using GPR [J]. Proc. of SPIE , 2005, 5794: 1233-1244.
[14] Geng, N. and Carin, L., Resonances of a dielectric BOR buried in a lossy,dispersive layered medium [C], International Geoscience and Remote Sensing Symposium (IGARSS), Seattle, WA, USA, 1998:222-224
[15] Casey, K. F. and Oetzel, G. N., Physical-optics models for scatterng from bured mines [C], Second Australian-American Joint Conference on the Technology of Mine Countermeasures, Sydney, Australia, 2001:1-15
[16] Andrieu J, Gallais F, and Mallepeyre V, et al. Land mine detection with an ultra-wideband SAR system [J]. Proceedings of SPIE, 2002, 4742:237-247.
[17] Sun Y J and Li J. Landmine detection using forward-looking ground penetrating radar [J]. Proceedings of SPIE, 2005, 5794: 1089-1097.
[18] Sun Y J and Li J. Adaptive learning approach to landmine detection [J]. IEEE Transactions on Aerospace and Electronic Systems, 2005, 41(3): 973-985.
[19] Allen, M. R., Efficient approach to physics-based FOPEN-SAR ATD/R [J], Proceedings of SPIE, 1996, 2757:163-172
[20] Chaney, R. D., Willsky, A. S. and Novak, L. M., Coherent aspect-dependent SAR image formation [J], Proceedings of SPIE, 1994, 2230:256-274
[21] Allen, M. R., Jauregui, J. M. and Hoff, L. E., FOPEN-SAR detection by direct use of simple scattering physics [C], IEEE National Radar Conference, Alexandria, VA, USA, 1995:152-157
[22]蒋咏梅,叶簇穿透超宽带SAR目标检测方法研究, [D],长沙:国防科学技术大学, 1998, 10.
[23] Koklnozi A, Hosgood B, and Sieber A J. Landmine detection using a time-sequence of thermal infrared images [J]. Proceedings of SPIE, 2003, 4885:170-179
[24] Cheng J and Miller E. Model-based principal component techniques for detection of buried landmines in multiframe synthetic aperture radar images. 2002:334-336
[25]王正明等, SAR图像提高分辨率技术[M],科学出版社,2006
[26]王鹏宇,超宽带机载步进频率SAR成像方法研究, [D],长沙:国防科学技术大学,2008, 11
[27]杨延光,周智敏,宋千.分裂孔径发射阵列接收BP算法方位分辨率分析[J].信号处理, 2008, 24(5):752-756
[28]王建,车载前视超宽带SAR浅埋目标成像技术研究, [D],长沙:国防科学技术大学, 2008, 10
[29]刘光平,超宽带合成孔径雷达高效成像算法, [D],长沙:国防科学技术大学, 2003, 03
[30]朱国富,董臻,梁甸农.超宽带LFM信号的BP成像算法[J].信号处理, 2001,17(5): 424-428
[31]聂在平等,目标与环境电磁散射特性建模[M],国防工业出版社, 2009
[32]文舸一.计算电磁学的进展与展望[J].电子学报, 1995,23(10):62-69
[33] K S Yee. Numerical solution of initial boundary value problems involving Maxwell’s equations in isotropic media [J]. IEEE Trans, Ap-14(4): 302-306,1966
[34]葛德彪等,电磁波时域有限差分方法[M],西安:西安电子科技大学出版社,2002
[35] Harrington R.F. Time-Harmonic Electromagnetic Fields. New York [DB]: McGraw-Hill, 1968
[36]孙晓坤,埋地地雷超宽带电磁散射特性及在目标检测中的应用, [D],长沙:国防科学技术大学, 2008, 7
[37] Kell R E. On the Derivation of Bistatic RCS from Monostatic Measurements [J]. Proc. of the IEEE, 1965(8): 983~988
[38] Sun Y J and Li J. Time-frequency analysis for plastic landmine detection via forward-looking ground penetrating radar [J]. IEE Proc.-Radar Sonar Navig, August 2003, 150(4): 253-261.
[39]杨延光,基于车载FLGPSAR序列图像的浅埋目标检测技术研究, [D],长沙:国防科学技术大学, 2008, 11
[40] Sinan B, John G. Unsupervised iterative detection of landmines in highly cluttered environments [J]. IEEE Transaction on Image Progressing, 2003, 12(5): 509-523
[41]王玉明,车载前视超宽带GPSAR杂波特性与目标特征研究, [D],长沙:国防科学技术大学,2008, 11
[42]杨志国,基于ROI的UWB SAR叶簇覆盖目标鉴别方法研究, [D],长沙:国防科学技术大学, 2007, 9
[43]谢冰,贺志毅.雷达导引头真实地杂波数据分析[J].制导与引信,2007, 28(1): 7-10,24
[44]王小鹏等译,形态学图像分析原理与应用[M],清华大学出版社, 2008
[45] Vitebskiy S and Carin L Short-pulse scattering from buried wires and bodies of revolution [C]. Proceedings of SPIE, 2002, 2747:233-244
[46] Jin Tian, Zhou Zhi-min, Song Qian, Chang Wen-ge. A novel SVM for ground penetrating synthetic aperture radar landmine detection [J]. Journal Of Electronics (CHINA), 2008, 25(1): 70-75
[47] Li J. and Stoical P. Efficient An adaptive filtering approach to spectral estimation and SAR imaging [J]. IEEE Trans on signal processing, Vol.44, No.6, 1996, pp. 1469:1484
[48] Hongbin Li, Jian Li, Petre Stoica, Performance analysis of forward-backward matched-filterbank spectral estimators [J], IEEE Transactions on Signal Processing, 1998, 46(7):1954-1966
[49]任璐,关于提高SAR图像质量的研究, [D],西安电子科技大学,2005, 1
[50]金添,宋千,孙晓坤,周智敏.地表穿透合成孔径雷达浅地表金属地雷二维电磁特征研究[J],电子学报, 2006, 34(12):2246-2249
[51] Donoho D L. Compressed sensing [J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289?1306
[52] Candes E, Tao T, Decoding by linear programming [J], IEEE Transactions on Information Theory, 2005, 51(12):4203-4215
[53] Chen S B, Donoho D L, Saunders M A. Atomic decomposition by basis pursuit [J]. SIAM Journal on Scientific Computing, 1998, 20(1): 33-61
[54]黄琼,屈乐乐,吴秉横,方广有.压缩感知在超宽带雷达成像中的应用.电波科学学报[J], 2010, 25(1):77-80
[55] Candes E and Tao T. The dantzig selector: statistical estimation when p is much larger than n [J]. The Annals of Statistics, 2007, 35(6): 1-41