基于极化SAR梯度和复Wishart分类器的舰船检测

殷君君 罗嘉豪 李响 代晓康 杨健

殷君君, 罗嘉豪, 李响, 等. 基于极化SAR梯度和复Wishart分类器的舰船检测[J]. 雷达学报(中英文), 2024, 13(2): 396–397. doi: 10.12000/JR23198
引用本文: 殷君君, 罗嘉豪, 李响, 等. 基于极化SAR梯度和复Wishart分类器的舰船检测[J]. 雷达学报(中英文), 2024, 13(2): 396–397. doi: 10.12000/JR23198
YIN Junjun, LUO Jiahao, LI Xiang, et al. Ship detection based on polarimetric SAR gradient and complex Wishart classifier[J]. Journal of Radars, 2024, 13(2): 396–397. doi: 10.12000/JR23198
Citation: YIN Junjun, LUO Jiahao, LI Xiang, et al. Ship detection based on polarimetric SAR gradient and complex Wishart classifier[J]. Journal of Radars, 2024, 13(2): 396–397. doi: 10.12000/JR23198

基于极化SAR梯度和复Wishart分类器的舰船检测

doi: 10.12000/JR23198
基金项目: 国家自然科学基金(62222102, 62171023),中央高校基本科研业务费(FRF-TP-22-005C1)
详细信息
    作者简介:

    殷君君,博士,教授,博士生导师,主要研究方向为极化雷达应用基础理论、极化合成孔径雷达图像理解、海洋遥感及生态环境变化监测等

    罗嘉豪,硕士生,主要研究方向为极化合成孔径雷达图像理解与应用、海面舰船检测、雷达目标检测和深度学习

    李 响,博士,高级工程师,主要研究方向为雷达目标检测和识别

    代晓康,硕士,主要研究方向为极化合成孔径雷达图像理解、海面舰船检测、地面车辆检测

    杨 健,博士,教授,博士生导师,主要研究方向为极化雷达理论及其应用等

    通讯作者:

    殷君君 junjun_yin@ustb.edu.cn

  • 责任主编:刘涛 Corresponding Editor: LIU Tao
  • 中图分类号: TN958

Ship Detection Based on Polarimetric SAR Gradient and Complex Wishart Classifier

Funds: The National Natural Science Foundation of China (62222102, 62171023), Fundamental Research Funds for the Central Universities (FRF-TP-22-005C1)
More Information
  • 摘要: 舰船检测是极化SAR系统的重要应用之一。现有的舰船检测方法容易受到旁瓣泄露的干扰,使得舰船目标的形态难以提取,导致检测结果不符合真实情况。此外,在舰船过于密集、尺度不一致的情况下,相邻舰船由于旁瓣的影响有时会被认为是单个目标,从而造成漏检。针对这些问题,该文提出一种基于极化SAR梯度和复Wishart分类器的舰船检测方法。首先,将似然比检验(LRT)梯度引入对数比值梯度框架,使其适用于极化SAR数据;基于LRT梯度图进行恒虚警(CFAR)检测,提取舰船的边缘信息,消除伪影的同时抑制强旁瓣对舰船精细轮廓提取的影响。其次,利用复Wishart迭代分类器对舰船强散射部分进行检测,可排除大部分的杂波干扰且保持舰船形态细节。最后,将二者信息融合,从而可以保持舰船形态细节的同时克服旁瓣和伪信号的虚警。该文在3幅来自ALOS-2卫星的极化SAR图像上进行了对比实验,实验表明与其他方法相比,该文所提算法具有更少的虚警和漏检,且能够有效克服旁瓣泄露,保持舰船形态细节。

     

  • 图  1  ROA示意图

    Figure  1.  Scheme of the ROA method

    图  2  所提算法流程图

    Figure  2.  Flow chart of the proposed method

    图  3  实验场景1

    Figure  3.  Experimental scene 1

    图  4  实验场景2 (第1行)和实验场景3 (第2行)

    Figure  4.  Experimental scene 2 (first line) and experimental scene 3 (second line)

    图  5  梯度结果

    Figure  5.  Calculation results for different gradients

    图  6  3种LRT梯度计算结果

    Figure  6.  Comparison of three LRT gradient calculation results

    图  7  LRT梯度的检测结果

    Figure  7.  The detection result of LRT gradient

    图  8  PWF的检测结果

    Figure  8.  The detection result of PWF

    图  9  复Wishart分类器的分类结果

    Figure  9.  The classification result of complex Wishart classifier

    图  10  LRT结果与复Wishart结果相加的伪彩色图

    Figure  10.  Pseudo-color display of the sum of LRT detection result and complex Wishart result

    图  11  融合检测结果

    Figure  11.  The fused result

    图  12  伪影和旁瓣示意图

    Figure  12.  Image of pseudo-signal and sidelobes

    图  13  基于LRT的CFAR检测结果

    Figure  13.  CFAR detection results based on LRT

    图  14  复杂海面融合检测结果

    Figure  14.  The fused result for complex sea surface scenario

    图  15  多尺度舰船检测结果

    Figure  15.  Multi-scale ship detection results

    图  16  实验场景2舰船检测结果

    Figure  16.  Ship detection results for experimental scenario 2

    图  17  实验场景3舰船检测结果

    Figure  17.  Ship detection results for experimental scenario 3

    图  18  实验场景1中的舰船检测结果

    Figure  18.  Ship detection results for experimental scenario 1

    表  3  实验场景1中对比方法的指标评价结果

    Table  3.   Results evaluation for experimental scenario 1

    方法召回率准确率F1时间(s)
    OS-CFAR0.41740.53860.47031.548
    Span0.44300.87150.58740.142
    PMS0.45170.86310.59310.152
    2P-CFAR0.46600.84940.601817.339
    Parzen0.46940.84770.60420.145
    T110.47370.93020.62770.138
    CA-CFAR0.67930.85660.75774.325
    λ30.71160.89810.794032.468
    OPCE0.71670.89170.794719.845
    GOPCE0.71670.89170.794729.904
    本文方法0.74710.91770.823792.673
    下载: 导出CSV

    表  2  实验场景3中对比方法的指标评价结果

    Table  2.   Results evaluation for experimental scenario 3

    方法虚警漏检召回率准确率F1时间(s)
    OS-CFAR2030.9690.8230.8905.765
    2P-CFAR1401.0000.8760.93461.898
    Span1210.9900.8900.9370.283
    CA-CFAR1110.9900.8980.94215.530
    Parzen1110.9900.8980.9420.261
    T11820.9790.9220.9500.248
    PSH1001.0000.9080.95290.842
    PMS630.9690.9390.9540.314
    λ3310.9900.9700.98091.453
    OPCE110.9900.9900.99067.250
    GOPCE110.9900.9900.99088.925
    本文方法010.9901.0000.995276.324
    下载: 导出CSV

    表  1  实验场景2中对比方法的指标评价结果

    Table  1.   Results evaluation for experimental scenario 2

    方法虚警漏检召回率准确率F1时间(s)
    PSH1040.8950.7730.82992.245
    2P-CFAR250.8650.9410.90161.870
    CA-CFAR250.8650.9410.90115.420
    Span430.9230.9000.9110.283
    Parzen420.9500.9050.9270.254
    T11510.9760.8890.9300.252
    OS-CFAR230.9230.9470.9355.981
    PMS320.9500.9270.9380.322
    OPCE120.9500.9740.96269.524
    λ3210.9760.9520.96491.350
    GOPCE020.9501.0000.97489.484
    本文方法001.0001.0001.000298.198
    下载: 导出CSV
  • [1] 杨汝良, 戴博伟, 李海英. 极化合成孔径雷达极化层次和系统工作方式[J]. 雷达学报, 2016, 5(2): 132–142. doi: 10.12000/JR16013.

    YANG Ruliang, DAI Bowei, and LI Haiying. Polarization hierarchy and system operating architecture for polarimetric synthetic aperture radar[J]. Journal of Radars, 2016, 5(2): 132–142. doi: 10.12000/JR16013.
    [2] WEISS M. Analysis of some modified cell-averaging CFAR processors in multiple-target situations[J]. IEEE Transactions on Aerospace and Electronic Systems, 1982, AES-18(1): 102–114. doi: 10.1109/TAES.1982.309210.
    [3] NOVAK L M, BURL M C, and IRVING W W. Optimal polarimetric processing for enhanced target detection[J]. IEEE Transactions on Aerospace and Electronic Systems, 1993, 29(1): 234–244. doi: 10.1109/7.249129.
    [4] AI Jiaqiu, YANG Xuezhi, DONG Zhangyu, et al. A new two parameter CFAR ship detector in Log-Normal clutter[C]. 2017 IEEE Radar Conference, Seattle, USA, 2017: 195–199. doi: 10.1109/RADAR.2017.7944196.
    [5] RITCEY J A and DU H. Order statistic CFAR detectors for speckled area targets in SAR[C]. Conference Record of the Twenty-Fifth Asilomar Conference on Signals, Systems & Computers, Pacific Grove, USA, 1991: 1082–1086. doi: 10.1109/ACSSC.1991.186613.
    [6] GOLDSTEIN G B. False-alarm regulation in log-normal and Weibull clutter[J]. IEEE Transactions on Aerospace and Electronic Systems, 1973, AES-9(1): 84–92. doi: 10.1109/TAES.1973.309705.
    [7] LEVANON N and SHOR M. Order statistcs CFAR for Weibull background[J]. IEE Proceedings F (Radar and Signal Processing), 1990, 137(3): 157–162. doi: 10.1049/ip-f-2.1990.0023.
    [8] ANASTASSOPOULOS V and LAMPROPOULOS G A. Optimal CFAR detection in Weibull clutter[J]. IEEE Transactions on Aerospace and Electronic Systems, 1995, 31(1): 52–64. doi: 10.1109/7.366292.
    [9] GUAN Jian, HE You, and PENG Yingning. CFAR detection in K-distributed clutter[C]. Fourth International Conference on Signal Processing, Beijing, China, 1998: 1513–1516. doi: 10.1109/ICOSP.1998.770911.
    [10] NOVAK L M and BURL M C. Optimal speckle reduction in polarimetric SAR imagery[J]. IEEE Transactions on Aerospace and Electronic Systems, 1990, 26(2): 293–305. doi: 10.1109/7.53442.
    [11] 杨汝良, 戴博伟, 谈璐璐, 等. 极化微波成像[M]. 北京: 国防工业出版社, 2016.

    YANG Ruliang, DAI Bowei, TAN Lulu, et al. Polarimetric Microwave Imaging[M]. Beijing: National Defense Industry Press, 2016.
    [12] CHANEY R D, BUD M C, and NOVAK L M. On the performance of polarimetric target detection algorithms[J]. IEEE Aerospace and Electronic Systems Magazine, 1990, 5(11): 10–15. doi: 10.1109/62.63157.
    [13] BOERNER W M, KOSTINSKI A B, and JAMES B D. On the concept of the polarimetric matched filter in high resolution radar imaging: An alternative for speckle reduction[C]. International Geoscience and Remote Sensing Symposium, ‘Remote Sensing: Moving Toward the 21st Century’, Edinburgh, UK, 1988: 69–72. doi: 10.1109/IGARSS.1988.570053.
    [14] KOSTINSKI A and BOERNER W. On the polarimetric contrast optimization[J]. IEEE Transactions on Antennas and Propagation, 1987, 35(8): 988–991. doi: 10.1109/TAP.1987.1144209.
    [15] YANG Jian, DONG Guiwei, PENG Yingning, et al. Generalized optimization of polarimetric contrast enhancement[J]. IEEE Geoscience and Remote Sensing Letters, 2004, 1(3): 171–174. doi: 10.1109/LGRS.2004.830127.
    [16] 殷君君, 安文韬, 杨健. 基于极化散射参数与Fisher-OPCE的监督目标分类[J]. 清华大学学报: 自然科学版, 2011, 51(12): 1782–1786. doi: 10.16511/j.cnki.qhdxxb.2011.12.007.

    YIN Junjun, AN Wentao, and YANG Jian. Supervised target classification using polarimetric scattering parameters and Fisher-OPCE[J]. Journal of Tsinghua University : Science and Technolog y, 2011, 51(12): 1782–1786. doi: 10.16511/j.cnki.qhdxxb.2011.12.007.
    [17] XI Yuyang, ZHANG Xi, LAI Quan, et al. A new PolSAR ship detection metric fused by polarimetric similarity and the third eigenvalue of the coherency matrix[C]. 2016 IEEE International Geoscience and Remote Sensing Symposium, Beijing, China, 2016: 112–115. doi: 10.1109/IGARSS.2016.7729019.
    [18] ZHANG Tao, YANG Zhen, and XIONG Huilin. PolSAR ship detection based on the polarimetric covariance difference matrix[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2017, 10(7): 3348–3359. doi: 10.1109/JSTARS.2017.2671904.
    [19] SHI Hao, ZHANG Qingjun, BIAN Mingming, et al. A novel ship detection method based on gradient and integral feature for single-polarization synthetic aperture radar imagery[J]. Sensors, 2018, 18(2): 563. doi: 10.3390/s18020 563.
    [20] DELLINGER F, DELON J, GOUSSEAU Y, et al. SAR-SIFT: A SIFT-like algorithm for SAR images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(1): 453–466. doi: 10.1109/TGRS.2014.2323552.
    [21] SONG Shengli, XU Bin, and YANG Jian. SAR target recognition via supervised discriminative dictionary learning and sparse representation of the SAR-HOG feature[J]. Remote Sensing, 2016, 8(8): 683. doi: 10.3390/rs8080683.
    [22] LIN Huiping, SONG Shengli, and YANG Jian. Ship classification based on MSHOG feature and task-driven dictionary learning with structured incoherent constraints in SAR images[J]. Remote Sensing, 2018, 10(2): 190. doi: 10.3390/rs10020190.
    [23] GAO Gui. A parzen-window-kernel-based CFAR algorithm for ship detection in SAR images[J]. IEEE Geoscience and Remote Sensing Letters, 2011, 8(3): 557–561. doi: 10.1109/LGRS.2010.2090492.
    [24] 张晓玲, 张天文, 师君, 等. 基于深度分离卷积神经网络的高速高精度SAR舰船检测[J]. 雷达学报, 2019, 8(6): 841–851. doi: 10.12000/JR19111.

    ZHANG Xiaoling, ZHANG Tianwen, SHI Jun, et al. High-speed and high-accurate SAR ship detection based on a depthwise separable convolution neural network[J]. Journal of Radars, 2019, 8(6): 841–851. doi: 10.12000/JR19111.
    [25] FAN Weiwei, ZHOU Feng, BAI Xueru, et al. Ship detection using deep convolutional neural networks for PolSAR images[J]. Remote Sensing, 2019, 11(23): 2862. doi: 10.3390/rs11232862.
    [26] AN Quanzhi, PAN Zongxu, and YOU Hongjian. Ship detection in Gaofen-3 SAR images based on sea clutter distribution analysis and deep convolutional neural network[J]. Sensors, 2018, 18(2): 334. doi: 10.3390/s18020334.
    [27] ZHANG Zhimian, WANG Haipeng, XU Feng, et al. Complex-valued convolutional neural network and its application in polarimetric SAR image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(12): 7177–7188. doi: 10.1109/TGRS.2017.2743222.
    [28] RIGNOT E J M and VAN ZYL J J. Change detection techniques for ERS-1 SAR data[J]. IEEE Transactions on Geoscience and Remote Sensing, 1993, 31(4): 896–906. doi: 10.1109/36.239913.
    [29] MA Xiaoshuang, SHEN Huanfeng, ZHANG Liangpei, et al. Adaptive anisotropic diffusion method for polarimetric SAR speckle filtering[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(3): 1041–1050. doi: 10.1109/JSTARS.2014.2328332.
    [30] NIELSEN A A, CONRADSEN K, and SKRIVER H. Change detection in full and dual polarization, single-and multifrequency SAR data[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(8): 4041–4048. doi: 10.1109/JSTARS.2015.2416434.
    [31] LEE J S, GRUNES M R, AINSWORTH T L, et al. Unsupervised classification using polarimetric decomposition and the complex Wishart classifier[J]. IEEE Transactions on Geoscience and Remote Sensing, 1999, 37(5): 2249–2258. doi: 10.1109/36.789621.
    [32] 张嘉峰, 朱博, 张鹏, 等. Wishart分布情形下极化SAR图像目标CFAR检测解析方法[J]. 电子学报, 2018, 46(2): 433–439. doi: 10.3969/j.issn.0372-2112.2018.02.024.

    ZHANG Jiafeng, ZHU Bo, ZHANG Peng, et al. Polarimetric SAR imagery target CFAR detection analytical algorithm with Wishart distribution[J]. Acta Electronica Sinica, 2018, 46(2): 433–439. doi: 10.3969/j.issn.0372-2112.2018.02.024.
    [33] WANG Hongmiao, ZENG Liang, ZHANG Tao, et al. A PolSAR despeckling method based on Wishart gradient and anisotropic diffusion[J]. Electronics Letters, 2021, 57(3): 126–128. doi: 10.1049/ell2.12086.
  • 加载中
图(18) / 表(3)
计量
  • 文章访问数:  401
  • HTML全文浏览量:  91
  • PDF下载量:  118
  • 被引次数: 0
出版历程
  • 收稿日期:  2023-10-10
  • 修回日期:  2023-11-15
  • 网络出版日期:  2023-12-07
  • 刊出日期:  2024-04-28

目录

    /

    返回文章
    返回