改进的L1/2阈值迭代高分辨率SAR成像算法

高志奇 孙书辰 黄平平 乞耀龙 徐伟

高志奇, 孙书辰, 黄平平, 等. 改进的L1/2阈值迭代高分辨率SAR成像算法[J]. 雷达学报, 2023, 12(5): 1044–1055. doi: 10.12000/JR22243
引用本文: 高志奇, 孙书辰, 黄平平, 等. 改进的L1/2阈值迭代高分辨率SAR成像算法[J]. 雷达学报, 2023, 12(5): 1044–1055. doi: 10.12000/JR22243
GAO Zhiqi, SUN Shuchen, HUANG Pingping, et al. Improved L1/2 threshold iterative high resolution SAR imaging algorithm[J]. Journal of Radars, 2023, 12(5): 1044–1055. doi: 10.12000/JR22243
Citation: GAO Zhiqi, SUN Shuchen, HUANG Pingping, et al. Improved L1/2 threshold iterative high resolution SAR imaging algorithm[J]. Journal of Radars, 2023, 12(5): 1044–1055. doi: 10.12000/JR22243

改进的L1/2阈值迭代高分辨率SAR成像算法

doi: 10.12000/JR22243
基金项目: 国家自然科学基金(61761037, 62071258),内蒙古自治区自然科学基金(2021MS06005, 2020ZD18),内蒙古自治区直属高校基本科研业务费项目(JY20220147)
详细信息
    作者简介:

    高志奇,博士,副教授,主要研究方向为阵列信号处理、SAR成像

    孙书辰,硕士生,主要研究方向为SAR成像算法

    黄平平,博士,教授,主要研究方向为新体制雷达系统、雷达信号处理和微波遥感应用

    乞耀龙,博士,副教授,主要研究方向为阵列成像、地基合成孔径雷达系统

    徐 伟,博士,教授,主要研究方向为新体制星载SAR系统仿真与信号处理

    通讯作者:

    黄平平 hpp@imut.edu.cn

  • 责任主编:张群 Corresponding Editor: ZHANG Qun
  • 中图分类号: TN957.5

Improved L1/2 Threshold Iterative High Resolution SAR Imaging Algorithm

Funds: The National Natural Science Foundation of China (61761037, 62071258), The Natural Science Foundation of Inner Mongolia (2021MS06005, 2020ZD18), Basic Scientific Research Business Cost Project of Colleges Directly under the Inner Mongolia (JY20220147)
More Information
  • 摘要: 针对合成孔径雷达(SAR)在稀疏采样条件下方位向分辨率低、易受噪声干扰等问题,提出改进的高分辨率SAR成像算法。该文在现有的L1/2正则化理论及其阈值迭代算法的基础上,改进了其表达式中的梯度算子,提高重构图像的求解精度,降低计算量。然后,在全采样和欠采样条件下,将原有L1/2阈值迭代算法与所提改进L1/2阈值迭代算法,分别结合近似观测模型对SAR回波信号进行成像处理和性能对比。实验结果表明,改进的算法具有更加优越的收敛性能,并且对于SAR图像方位向分辨率有一定的改善。

     

  • 图  1  条带式SAR空间模型图

    Figure  1.  Strip SAR spatial model diagram

    图  2  点目标位置

    Figure  2.  Point target position

    图  3  Chirp-Scaling算法成像结果

    Figure  3.  Imaging results of Chirp-Scaling algorithm

    图  4  L1/2阈值迭代算法成像结果

    Figure  4.  Imaging results of L1/2 threshold iterative algorithm

    图  5  改进L1/2阈值迭代算法成像结果

    Figure  5.  Imaging results of improved L1/2 threshold iterative algorithm

    图  6  点目标剖面图

    Figure  6.  Cross section of point targets

    图  7  Chirp-Scaling算法成像结果

    Figure  7.  Imaging results of Chirp-Scaling algorithm

    图  8  L1/2阈值迭代算法成像结果

    Figure  8.  Imaging results of the L1/2 threshold iterative algorithm

    图  9  改进L1/2阈值迭代算法成像结果

    Figure  9.  Imaging results of the improved L1/2 threshold iterative algorithm

    图  10  不同信噪比下成像效果对比

    Figure  10.  Imaging results for different SNR

    图  11  不同算法的重建结果

    Figure  11.  Reconstruction results of different algorithms

    表  1  SAR成像仿真参数

    Table  1.   Simulation parameters of SAR imaging

    参数数值
    载频(GHz)5.3
    飞行速度(m/s)150
    脉冲宽度(μs)2.5
    距离向采样率(MHz)60
    方位向采样率(Hz)200
    平台离场景中心斜距(km)20
    斜视角(°)0
    距离向调频率(MHz/μs)20000
    下载: 导出CSV

    表  2  3种算法中5个点目标成像分辨率分析(m)

    Table  2.   Imaging resolution analysis of five targets in three algorithms (m)

    算法点目标1点目标2点目标3点目标4点目标5
    Chirp-Scaling算法1.61041.62251.62351.62351.6235
    L1/2阈值迭代算法1.14251.11431.14251.14251.1425
    改进L1/2阈值迭代算法0.93890.93820.96230.96230.9623
    下载: 导出CSV

    表  3  3种算法中方位向数据缺失60%时5个点目标的分辨率(m)

    Table  3.   The resolution of five targets with 60% azimuth data missed for three algorithms (m)

    算法点目标1点目标2点目标3点目标4点目标5
    Chirp-Scaling算法1.66191.65221.64841.64841.6484
    L1/2阈值迭代算法1.14941.10051.14941.14941.1494
    改进L1/2阈值迭代算法1.07591.05021.05991.05991.0599
    下载: 导出CSV

    表  4  3种算法中方位向数据缺失70%时5个点目标的分辨率(m)

    Table  4.   The resolution of five targets with 70% azimuth data missed for three algorithms (m)

    算法点目标1点目标2点目标3点目标4点目标5
    Chirp-Scaling算法1.67191.78011.77681.77681.7768
    L1/2阈值迭代算法1.16041.20541.16041.16041.1604
    改进L1/2阈值迭代算法1.10491.06191.06271.06271.0627
    下载: 导出CSV

    表  5  不同模型的运算量分析

    Table  5.   Calculation amount analysis of different models

    性能分析匹配滤波精确观测模型近似观测模型
    空间复杂度O(MN)O(M2N2)O(MN)
    时间复杂度O(MNlog2MN)O(IM2N2)O(IMNlog2MN)
    下载: 导出CSV

    表  6  不同采样率下两种算法重建点目标耗时时长(s)

    Table  6.   Time consuming of point target reconstruction by the two algorithms at different sampling rates (s)

    算法采样率100.0%采样率75.0%采样率50.0%采样率25.0%采样率12.5%
    L1/2阈值迭代算法98.56059274.14680949.48575125.46455413.557194
    改进L1/2阈值迭代算法53.30758740.52535426.61006913.7619237.354359
    下载: 导出CSV

    表  7  RADARSAT-1卫星SAR成像参数

    Table  7.   Parameters of RADARSAT-1 satellite SAR imaging

    参数数值
    载频(GHz)5.3
    飞行速度(m/s)7062
    脉冲宽度(μs)41.75
    距离向采样率(MHz)32317
    脉冲重复频率(Hz)125.7
    平台离场景中心斜距(km)988.65
    图像分辨率单元数(M×N)2048×3000
    距离向调频率(MHz/μs)5000
    成像场景大小(m)10000×12000
    下载: 导出CSV

    表  8  3种算法实测数据成像耗时时长

    Table  8.   Time consuming of measured data imaging by the three algorithms

    算法重建耗时时长(s)
    Chirp-Scaling算法3.143686
    L1/2阈值迭代算法14.176919
    改进L1/2阈值迭代算法10.074317
    下载: 导出CSV
  • [1] GLENTIS G, ZHAO Kexin, JAKOBSSON A, et al. Non-parametric high-resolution SAR imaging[J]. IEEE Transactions on Signal Processing, 2013, 61(7): 1614–1624. doi: 10.1109/TSP.2012.2232662
    [2] DONOHO D L. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289–1306. doi: 10.1109/TIT.2006.871582
    [3] 李清泉, 王欢. 基于稀疏表示理论的优化算法综述[J]. 测绘地理信息, 2019, 44(4): 1–9. doi: 10.14188/j.2095-6045.2019015

    LI Qingquan and WANG Huan. Sparse representation based optimization: A survey[J]. Journal of Geomatics, 2019, 44(4): 1–9. doi: 10.14188/j.2095-6045.2019015
    [4] NI Jiacheng, ZHANG Qun, LUO Ying, et al. Compressed sensing SAR imaging based on centralized sparse representation[J]. IEEE Sensors Journal, 2018, 18(12): 4920–4932. doi: 10.1109/JSEN.2018.2831921
    [5] JUNG D H, KIM H S, KIM C K, et al. Sparse scene recovery for high-resolution automobile FMCW SAR via scaled compressed sensing[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(12): 10136–10146. doi: 10.1109/TGRS.2019.2931626
    [6] 王天云, 刘冰, 魏强, 等. 压缩感知成像雷达研究进展[J]. 电光与控制, 2019, 26(7): 1–8. doi: 10.3969/j.issn.1671-637X.2019.07.001

    WANG Tianyun, LIU Bing, WEI Qiang, et al. A review on research progresses of compressed sensing imaging radar[J]. Electronics Optics &Control, 2019, 26(7): 1–8. doi: 10.3969/j.issn.1671-637X.2019.07.001
    [7] JIANG Hai, JIANG Chenglong, ZHANG Bingchen, et al. Experimental results of spaceborne stripmap SAR raw data imaging via compressed sensing[C]. 2011 IEEE CIE International Conference on Radar, Chengdu, China, 2011: 202–205.
    [8] ALONSO M T, LOPEZ-DEKKER M, and MALLORQUI J J. A novel strategy for radar imaging based on compressive sensing[J]. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(12): 4285–4295. doi: 10.1109/TGRS.2010.2051231
    [9] ZENG Jinshan, FANG Jian, and XU Zongben. Sparse SAR imaging based on L1/2 regularization[J]. Science China Information Sciences, 2012, 55(8): 1755–1775. doi: 10.1007/s11432-012-4632-5
    [10] 史洪印, 贾宝京, 齐兆龙. 基于压缩感知的非均匀脉冲SAR欺骗性干扰抑制方法[J]. 仪器仪表学报, 2016, 37(3): 525–532. doi: 10.3969/j.issn.0254-3087.2016.03.007

    SHI Hongyin, JIA Baojing, and QI Zhaolong. Novel non-uniform pulse SAR deception jamming suppressing method based on compressive sensing[J]. Chinese Journal of Scientific Instrument, 2016, 37(3): 525–532. doi: 10.3969/j.issn.0254-3087.2016.03.007
    [11] 段化军, 朱岱寅, 李勇, 等. 基于压缩感知的条带SAR缺失数据恢复成像方法[J]. 系统工程与电子技术, 2016, 38(5): 1025–1031. doi: 10.3969/j.issn.1001-506X.2016.05.09

    DUAN Huajun, ZHU Daiyin, LI Yong, et al. Recovery and imaging method for missing data of the strip-map SAR based on compressive sensing[J]. Systems Engineering and Electronics, 2016, 38(5): 1025–1031. doi: 10.3969/j.issn.1001-506X.2016.05.09
    [12] 李博, 刘发林, 周崇彬, 等. 基于近似观测的加权L1压缩感知SAR成像[J]. 微波学报, 2018, 34(6): 62–67. doi: 10.14183/j.cnki.1005-6122.201806014

    LI Bo, LIU Falin, ZHOU Chongbin, et al. Approximated observation-based weighted L1 compressed sensing SAR imaging[J]. Journal of Microwaves, 2018, 34(6): 62–67. doi: 10.14183/j.cnki.1005-6122.201806014
    [13] 徐宗本, 吴一戎, 张冰尘, 等. 基于L1/2正则化理论的稀疏雷达成像[J]. 科学通报, 2018, 63(14): 1306–1319. doi: 10.1360/N972018-00372

    XU Zongben, WU Yirong, ZHANG Bingchen, et al. Sparse radar imaging based on L1/2 regularization theory[J]. Chinese Science Bulletin, 2018, 63(14): 1306–1319. doi: 10.1360/N972018-00372
    [14] 杨卫星, 朱岱寅. 稀疏场景下SAR方位向随机丢失数据的迭代成像算法[J]. 系统工程与电子技术, 2021, 43(7): 1748–1755. doi: 10.12305/j.issn.1001-506X.2021.07.03

    YANG Weixing and ZHU Daiyin. Iterative imaging algorithm for SAR azimuth random missing data with sparse scenes[J]. Systems Engineering and Electronics, 2021, 43(7): 1748–1755. doi: 10.12305/j.issn.1001-506X.2021.07.03
    [15] ZHANG Jian and GHANEM B. ISTA-net: Interpretable optimization-inspired deep network for image compressive sensing[C]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 1828–1837.
    [16] 赵克祥, 毕辉, 张冰尘. 基于快速阈值迭代的SAR层析成像处理方法[J]. 系统工程与电子技术, 2017, 39(5): 1019–1023. doi: 10.3969/j.issn.1001-506X.2017.05.11

    ZHAO Kexiang, BI Hui, and ZHANG Bingchen. SAR tomography method based on fast threshold iteration iterative shrinkage-thresholding[J]. Systems Engineering and Electronics, 2017, 39(5): 1019–1023. doi: 10.3969/j.issn.1001-506X.2017.05.11
    [17] BI Hui and BI Guoan. Performance analysis of iterative soft thresholding algorithm for L1 regularization based sparse SAR imaging[C]. 2019 IEEE Radar Conference, Boston, USA, 2019: 1–6.
    [18] XU Zhongming, WANG Qinghua, HE Yansong, et al. A monotonic two-step iterative shrinkage/thresholding algorithm for sound source identification based on equivalent source method[J]. Applied Acoustics, 2018, 129: 386–396. doi: 10.1016/j.apacoust.2017.07.012
    [19] NESTEROV Y E. A method for solving the convex programming problem with convergence rate O (1/k2)[J]. Doklady Akademii Nauk SSSR, 1983, 269(3): 543–547.
    [20] BECK A and TEBOULLE M. A fast iterative shrinkage-thresholding algorithm for linear inverse problems[J]. SIAM Journal on Imaging Sciences, 2009, 2(1): 183–202. doi: 10.1137/080716542
    [21] 张倩, 李海洋. 一种改进的迭代软阈值算法及其应用[J]. 纺织高校基础科学学报, 2018, 31(2): 253–260. doi: 10.13338/j.issn.1006-8341.2018.02.020

    ZHANG Qian and LI Haiyang. An improved iterative soft thresholding algorithm and application[J]. Basic Sciences Journal of Textile Universities, 2018, 31(2): 253–260. doi: 10.13338/j.issn.1006-8341.2018.02.020
    [22] XU Zongben. Data modeling: Visual psychology approach and L1/2 regularization theory[C]. International Congress of Mathematicians 2010 (ICM 2010), Hyderabad, India, 2010: 3151–3184.
    [23] XU Zongben, GUO Hailiang, WANG Yao, et al. Representative of L1/2 regularization among Lq ( 0<q≤ 1) regularizations: An experimental study based on phase diagram[J]. Acta Automatica Sinica, 2012, 38(7): 1225–1228. doi: 10.1016/S1874-1029(11)60293-0
    [24] XU Zongben, CHANG Xiangyu, XU Fengmin, et al. L1/2 regularization: A thresholding representation theory and a fast solver[J]. IEEE Transactions on Neural Networks and Learning Systems, 2012, 23(7): 1013–1027. doi: 10.1109/TNNLS.2012.2197412
    [25] ZENG Jinshan, LIN Shaobo, WANG Yao, et al. L1/2 regularization: Convergence of iterative half thresholding algorithm[J]. IEEE Transactions on Signal Processing, 2014, 62(9): 2317–2329. doi: 10.1109/TSP.2014.2309076
    [26] BI Hui, ZHU Daiyin, BI Guoan, et al. FMCW SAR sparse imaging based on approximated observation: An overview on current technologies[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13: 4825–4835. doi: 10.1109/JSTARS.2020.3017487
    [27] FANG Jian, XU Zongben, ZHANG Bingchen, et al. Fast compressed sensing SAR imaging based on approximated observation[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(1): 352–363. doi: 10.1109/JSTARS.2013.2263309
  • 加载中
图(11) / 表(8)
计量
  • 文章访问数:  543
  • HTML全文浏览量:  331
  • PDF下载量:  197
  • 被引次数: 0
出版历程
  • 收稿日期:  2022-12-28
  • 修回日期:  2023-02-05
  • 网络出版日期:  2023-02-22
  • 刊出日期:  2023-10-28

目录

    AI智能问答

        /

        返回文章
        返回