基于形态学自适应分块的高分辨SAR多特征增强算法

doi:10.12305/j.issn.1001-506X.2022.02.15

摘要/Abstract

摘要：

传统基于$\ell_1 $范数正则化算子(least absolute shrinkage and selection operator，LASSO)模型的合成孔径雷达(synthetic aperture radar，SAR)压缩感知类稀疏成像算法易丢失弱散射点。基于扩展型组LASSO系列模型的算法虽可增强SAR结构特征以保留弱散射点，但由于其增强过程中本质上采用了欧式距离方法进行特征分块，致使分块较为“机械”, 不能很好地提取目标结构，从而影响最终高分辨SAR成像质量。针对上述问题，提出一种基于形态学自适应分块的交替方向多乘子法(morphological auto-blocking alternating direction method of multipliers，MAB-ADMM)来实现高分辨SAR多特征表征。该算法通过建立基于形态学分块的$\ell_{\rm{M}} $/$\ell_{\rm{F}} $混合结构范数和$\ell_1 $稀疏范数来分别引入结构和稀疏先验，从而实现结构与稀疏多特征增强。由于采用了基于测地距离的形态学分块方式，MAB-ADMM算法能够更加有效地识别感兴趣的目标轮廓，从而提高结构增强的准确度和完整度。实验部分通过采用仿真复数据和实测SAR数据对所提算法和传统算法的成像结果进行定性对比，从而验证所提算法具有优越的多特征增强能力。此外，采用相变热力图对所提算法和传统算法的恢复能力进行定量对比，并利用MSTAR数据验证了所提算法可有效针对分类算法进行结果提升。

关键词: 交替方向多乘子法, 合成孔径雷达成像, 形态学, 相变图

Abstract:

The conventional compressive sensing sparse imaging algorithm of synthetic aperture radar (SAR) based on the least absolute shrinkage and selection operator (LASSO) is easy to lose weak scatters. Although the imaging algorithm based on extended group LASSO is capable of strengthening weak scatters of structure of target of interests, it is inflexible to adaptively block the groups for enhancing the target feature, since only the Euclidean distance is adopted for the blocking operation, and the quality of the SAR imagery is poor. In this paper, a morphological auto-blocking alternating direction method of multipliers (MAB-ADMM) algorithm is proposed for the structural feature enhancement of high-resolution SAR imagery. A mixture of $\ell_{\rm{M}} $/$\ell_{\rm{F}} $ norm and an $\ell_1 $ norm are introduced to represent the structural and sparse features, respectively, and a coordination process is carried out to enhance the features cooperatively. Because the proposed MAB-ADMM algorithm employs the morphological blocking scheme based on the geodesic distance, it is capable of capturing the contour of target of interests effectively, and the accuracy and integrity of the enhancement can be improved. In the experiment, both the simulated and raw SAR data are used for the verification of the proposed algorithm. Comparisons with the conventional algorithm are performed to show the superiority of the proposed algorithm. The phase transition analysis is carried out to evaluate the performance of the proposed algorithm quantitatively. Also, the MSTAR data set is employed to verify the performance of the proposed algorithm in preparing for the SAR target classification.

Key words: alternating direction method of multipliers (ADMM), synthetic aperture radar (SAR) imaging, morphology, phase transition diagram (PTD)

中图分类号:

TN957.52

方澄, 李慧娟, 路稳, 宋玉蒙, 杨磊. 基于形态学自适应分块的高分辨SAR多特征增强算法[J]. 系统工程与电子技术, 2022, 44(2): 470-479.

Cheng FANG, Huijuan LI, Wen LU, Yumeng SONG, Lei YANG. Multi-feature enhancement algorithm for high resolution SAR based on morphological auto-blocking[J]. Systems Engineering and Electronics, 2022, 44(2): 470-479.

图/表 7

图1

图2

图3

图4

图5

图6

表1

参考文献 30

1	张璘, 姜义成. 基于速度合成孔径雷达的海面舰船动目标成像方法[J]. 系统工程与电子技术, 2020, 42 (1): 45- 51.
	ZHANG L , JIANG Y C . Imaging of moving surface ships based on velocity synthetic aperture radar[J]. Systems Engineering and Electronics, 2020, 42 (1): 45- 51.
2	LIU A F , WANG F , XU H , et al. N-SAR: a new multichannel multimode polarimetric airborne SAR[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2018, 11 (9): 3155- 3166. doi: 10.1109/JSTARS.2018.2848945
3	ZHANG Y , XIONG W , DONG X C , et al. A novel azimuth spectrum reconstruction and imaging method for moving targets in geosynchronous spaceborne-airborne bistatic multichannel SAR[J]. IEEE Trans.on Geoscience and Remote Sensing, 2020, 58 (8): 5976- 5991. doi: 10.1109/TGRS.2020.2974531
4	CHEN L P , AN D X , HUANG X T . Extended autofocus backprojection algorithm for low-frequency SAR imaging[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14 (8): 1323- 1327. doi: 10.1109/LGRS.2017.2711005
5	KANG M S , KIM K T . Ground moving target imaging based on compressive sensing framework with single-channel SAR[J]. IEEE Sensors Journal, 2020, 20 (3): 1238- 1250. doi: 10.1109/JSEN.2019.2947114
6	PENG X M , TAN W X , HONG W , et al. Airborne DLSLA 3-D SAR image reconstruction by combination of polar formatting and L1 regularization[J]. IEEE Trans.on Geoscience and Remote Sensing, 2016, 54 (1): 213- 226. doi: 10.1109/TGRS.2015.2453202
7	蒋留兵, 黄韬, 沈翰宁, 等. 基于局部随机化哈达玛矩阵的正交多匹配追踪算法[J]. 系统工程与电子技术, 2013, 35 (5): 914- 919. doi: 10.3969/j.issn.1001-506X.2013.05.03
	JIANG L B , HUANG T , SHEN H N , et al. Orthogonal multi-matching pursuit algorithm based on local randomized Hadamard matrix[J]. Systems Engineering and Electronics, 2013, 35 (5): 914- 919. doi: 10.3969/j.issn.1001-506X.2013.05.03
8	杨磊, 岳云泽, 李埔丞, 等. 多地面运动目标大动态SAR成像稀疏表示[J]. 西安电子科技大学学报, 2019, 46 (5): 31- 40.
	YANG L , YUE Y Z , LI P C , et al. Sparse representation of multi-ground moving targets in large dynamic SAR imaging[J]. Journal of Xidian University, 2019, 46 (5): 31- 40.
9	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.
	ZHAO K X , BI H , ZHANG B C . SAR tomography method based on fast iterative shrinkage-thresholding[J]. Systems Engineering and Electronics, 2017, 39 (5): 1019- 1023.
10	STEPHEN B , NEAL P , CHU E , et al. Distributed optimization and statistical learning via the alternating direction method of multipliers[J]. Foundations and Trends in Machine Learning, 2011, 3 (1): 1- 122.
11	杨磊, 李埔丞, 李慧娟, 等. 稳健高效通用SAR图像稀疏特征增强算法[J]. 电子与信息学报, 2019, 41 (12): 2826- 2835. doi: 10.11999/JEIT190173
	YANG L , LI P C , LI H J , et al. Robust and efficient general SAR image sparse feature enhancement algorithm[J]. Journal of Electronics and Information Technology, 2019, 41 (12): 2826- 2835. doi: 10.11999/JEIT190173
12	YUAN M , LIN Y . Model selection and estimation in regression with grouped variables[J]. Journal of the Royal Statistical Society Series B, 2006, 68, 49- 67. doi: 10.1111/j.1467-9868.2005.00532.x
13	杨磊, 李慧娟, 黄博, 等. 双层稀疏组LASSO高分辨SAR结构特征增强成像[J]. 系统工程与电子技术, 2021, 43 (2): 351- 362.
	YANG L , LI H J , HUANG B , et al. High resolution SAR imagery with structural feature enhancement under two-layer sparse group LASSO[J]. Systems Engineering and Electronics, 2021, 43 (2): 351- 362.
14	TANG C M , LIU X L , LI Y J , et al. Morphology operators construction by adaptive elliptical structuring elements based on nonlinear structure tensor[J]. Journal of Physics: Conference Series, 2017, 787 (1): 012021.
15	HARALICK R M , STERNBERG S R , ZHUANG X . Image analysis using mathematical morphology[J]. IEEE Trans.on Pattern Analysis and Machine Intelligence, 1987, 9 (4): 532- 550.
16	CHEN M , LIU D Y , QIAN K J , et al. Lunar crater detection based on terrain analysis and mathematical morphology methods using digital elevation models[J]. IEEE Trans.on Geoscience and Remote Sensing, 2018, 56 (7): 3681- 3692. doi: 10.1109/TGRS.2018.2806371
17	LIU L Y , JIA Z H , YANG J , et al. SAR image change detection based on mathematical morphology and the K-means clustering algorithm[J]. IEEE Access, 2019, 7, 43970- 43978. doi: 10.1109/ACCESS.2019.2908282
18	SGHAIER M O , FOUCHER S , LEPAGE R . River extraction from high-resolution SAR images combining a structural feature set and mathematical morphology[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2017, 10 (3): 1025- 1038. doi: 10.1109/JSTARS.2016.2609804
19	ZARANDY A , STOFFELS A , ROSKA T , et al. Implementation of binary and gray-scale mathematical morphology on the CNN universal machine[J]. IEEE Trans.on Circuits and Systems Ⅰ: Fundamental Theory and Applications, 1998, 45 (2): 163- 168. doi: 10.1109/81.661683
20	ZHANG A Q , JI T Y , LI M S , et al. An identification method based on mathematical morphology for sympathetic inrush[J]. IEEE Trans.on Power Delivery, 2018, 33 (1): 12- 21. doi: 10.1109/TPWRD.2016.2590479
21	ZHAO H M , LIU H D , XU J J , et al. Performance prediction using high-order differential mathematical morphology gradient spectrum entropy and extreme learning machine[J]. IEEE Trans.on Instrumentation and Measurement, 2020, 69 (7): 4165- 4172. doi: 10.1109/TIM.2019.2948414
22	ZHANG W B, WANG H J, TENG R J, et al. Application of adaptive structure element for generalized morphological filtering in vibratio signal de-noising[C]//Proc. of the 3rd International Congress on Image and Signal Processing, 2010: 3313-3317.
23	葛世国. 基于数学形态学的遥感图像分割算法研究[D]. 成都: 成都理工大学, 2014.
	GE S G. Research on remote sensing image segmentation algorithm based on mathematical morphology[D]. Chengdu: Chengdu University of Technology, 2014.
24	LU Z, WANG F L, CHANG Y Q, et al. Edge detection based on adaptive structure element morphology[C]//Proc. of the IEEE International Conference on Automation and Logistics, 2007: 254-257.
25	LI S, ZHOU G Q, ZHENG Z Z, et al. The relation between accuracy and size of structure element for vehicle detection with high resolution highway aerial images[C]//Proc. of the IEEE International Geoscience and Remote Sensing Symposium, 2013: 2645-2648.
26	YANG L , XING M D , WANG Y , et al. Compensation for the NsRCM and phase error after polar format resampling for airborne spotlight SAR raw data of high resolution[J]. IEEE Geoscience and Remote Sensing Letters, 2013, 10 (1): 165- 169. doi: 10.1109/LGRS.2012.2196676
27	YANG L , ZHAO L F , BI G A , et al. SAR ground moving target imaging algorithm based on parametric and dynamic sparse Bayesian learning[J]. IEEE Trans.on Geoscience and Remote Sensing, 2016, 54 (4): 2254- 2267. doi: 10.1109/TGRS.2015.2498158
28	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.
	LI Q Q , WANG H . A review of optimization algorithms based on sparse representation theory[J]. Journal of Geomatics, 2019, 44 (4): 1- 9.
29	DONOHO D L , ARIAN M , ANDREA M . The noise-sensitivity phase transition in compressed sensing[J]. IEEE Trans.on Information Theory, 2011, 57 (10): 6920- 6941. doi: 10.1109/TIT.2011.2165823
30	CHAN T H , JIA K , GAO S , et al. PCANet: a simple deep learning baseline for image classification[J]. IEEE Trans.on Image Processing, 2015, 24 (12): 5017- 5032. doi: 10.1109/TIP.2015.2475625

类别	分类方法
	PCANet算法		特征增强参数1		特征增强参数2		特征增强参数3		三组参数综合结果
	误分类数	准确率/%	误分类数	准确率/%	误分类数	准确率/%	误分类数	准确率/%	误分类数	准确率/%
SN9563	25	87.18	13	93.33	14	92.82	14	92.82	9	95.38
SN9566	47	76.02	42	78.57	37	81.12	33	83.16	28	85.71
SNC21	40	79.59	36	81.63	30	84.69	35	82.14	27	86.22
SNC71	0	100.0	0	100.0	0	100.0	0	100.0	0	100.0
SN132	29	85.20	22	87.76	20	89.80	19	90.31	12	93.88
SN812	13	93.33	6	96.92	8	95.90	8	95.90	4	97.95
SNS7	16	91.62	10	95.29	12	93.72	11	94.245	8	95.81
合计	170	87.55	129	90.55	121	91.14	120	91.21	88	93.55

[1]	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.
[2]	王曼颖, 龚晓峰, 雒瑞森, 边彤, 王之远. 基于自适应形态学的跳频信号参数联合盲估计[J]. 系统工程与电子技术, 2021, 43(5): 1398-1405.
[3]	杨磊, 李慧娟, 黄博, 刘伟, 李埔丞. 双层稀疏组Lasso高分辨SAR结构特征增强成像[J]. 系统工程与电子技术, 2021, 43(2): 351-362.
[4]	鲍雨婷, 曹菲, 张鹍鹏, 占必超, 李君宜, 占建伟. 基于OFDM-LFM的弹载广域SAR成像距离模糊抑制[J]. 系统工程与电子技术, 2021, 43(2): 369-375.
[5]	段佳, 曹兰英, 吴亿锋. 基于属性散射中心的SAR成像方法[J]. 系统工程与电子技术, 2021, 43(10): 2782-2788.
[6]	董淑仙, 全英汇, 陈侠达, 高霞, 李亚超, 邢孟道. 基于捷变频联合数学形态学的干扰抑制算法[J]. 系统工程与电子技术, 2020, 42(7): 1491-1498.
[7]	马彦恒, 李根, 熊旭颖, 侯建强. 机动平台大斜视压缩感知SAR成像方法[J]. 系统工程与电子技术, 2020, 42(10): 2197-2206.
[8]	王殿伟, 陈鹏, 李大湘, 刘颖, 许志杰, 王晶. 融合纹理信息的深度图像修复[J]. 系统工程与电子技术, 2019, 41(8): 1720-1725.
[9]	于欣永, 郭英, 张坤峰, 眭萍, 李雷, 李红光, 孟涛. 高效的多跳频信号2D-DOA估计算法[J]. 系统工程与电子技术, 2018, 40(6): 1363-1370.
[10]	张斌, 胡庆荣, 韦立登, 李爽. 改进的形态学干涉图滤波方法[J]. 系统工程与电子技术, 2018, 40(10): 2230-2236.
[11]	邹高翔, 童创明, 王童, 孙华龙. 地海交界分区域复合粗糙面建模及电磁散射特性研究[J]. 系统工程与电子技术, 2017, 39(7): 1425-1438.
[12]	张欢, 杨任农, 吴军, 李秋妮, 傅修竹, 孙长月. 多机协同电子战规划压制干扰布阵研究[J]. 系统工程与电子技术, 2017, 39(3): 542-548.
[13]	李军侠, 水鹏朗. 基于Canny振荡抑制准则的改进匹配滤波器[J]. 系统工程与电子技术, 2016, 38(7): 1543-1548.
[14]	毕辉, 蒋成龙, 王万影, 张冰尘, 洪文. 层析合成孔径雷达成像航迹分布优化方法[J]. 系统工程与电子技术, 2015, 37(8): 1787-1792.
[15]	李少东, 裴文炯, 杨军, 胡国旗. 贝叶斯模型下的OMP重构算法及应用[J]. 系统工程与电子技术, 2015, 37(2): 246-252.