基于混合稀疏表示的二维压缩感知SAR成像

熊世超; 倪嘉成; 张群; 罗迎; 王岩松

doi:10.13700/j.bh.1001-5965.2021.0101

基于混合稀疏表示的二维压缩感知SAR成像

doi: 10.13700/j.bh.1001-5965.2021.0101

熊世超¹,
倪嘉成^1, ,,
张群^{1, 2},
罗迎^{1, 2},
王岩松³

1.
空军工程大学信息与导航学院, 西安 710077
2.
复旦大学波散射与遥感信息国家教育部重点实验室, 上海 200433
3.
中国人民解放军 93303部队, 沈阳 110000

基金项目:

国家自然科学基金 62001508

国家自然科学基金 61871396

陕西省自然科学基础研究计划 2020JQ-480

陕西省自然科学基础研究计划 2020JM-348

详细信息

通讯作者:
倪嘉成, E-mail: littlenjc@sina.com

中图分类号: V243.2;TN957.5
计量
- 文章访问数: 262
- HTML全文浏览量: 87
- PDF下载量: 29
- 被引次数: 0
出版历程
- 收稿日期: 2021-03-02
- 录用日期: 2021-06-18
- 网络出版日期: 2021-07-07
- 整期出版日期: 2022-11-20

2-D compressed sensing SAR imaging based on mixed sparse representation

XIONG Shichao¹,
NI Jiacheng^{1
, ,},
ZHANG Qun^{1, 2},
LUO Ying^{1, 2},
WANG Yansong³

1.
Information and Navigation College, Air Force Engineering University, Xi'an 710077, China
2.
Key Laboratory of Wave Scattering and Remote Sensing Information, Fudan University, Shanghai 20043
3.
93303 Unit of PLA, Shenyang 110000, China

Funds:

National Natural Science Foundation of China 62001508

National Natural Science Foundation of China 61871396

Natural Science Basic Research Program of Shaanxi 2020JQ-480

Natural Science Basic Research Program of Shaanxi 2020JM-348

More Information

Corresponding author: NI Jiacheng, E-mail: littlenjc@sina.com

摘要

摘要:
压缩感知(CS)理论在合成孔径雷达(SAR)成像中应用广泛。针对包含城市、河流等区域的非稀疏场景压缩感知SAR成像, 提出基于近似观测模型的混合稀疏表示(MSR)压缩感知SAR成像方法。该方法将复杂的SAR图像分解成点、线、面, 并将线、面分别通过离散余弦变换和曲波变换转换到稀疏域, 使压缩感知的稀疏性条件得以满足, 通过求解基于近似观测模型的二维压缩感知优化问题重建非稀疏场景的SAR图像。所提方法能够实现降采样率条件下对包含城市、河流等非稀疏场景区域的成像, 仿真场景和实测场景成像结果表明了所提方法的有效性。
- 合成孔径雷达 /
- 压缩感知 /
- 稀疏表示 /
- 近似观测 /
- 曲波变换
Abstract:
Compressed sensing (CS) theory has been applied in synthetic aperture radar (SAR) in the recent years. A 2-D CS SAR imaging method is proposed using mixed sparse representation (MSR) based on approximate observation model in non-sparse scene compressed sensing SAR imaging. Firstly, non-sparse scene with complicated ground features is decomposed into point-like, edges and smooth components. Then, edges and smooth components are transformed into sparse domain by discrete cosine transform and curvelet transform respectively. And based on approximate observation model, SAR images are derived from 2-D CS optimization problem. Owing to the sparse representation method of non-sparse scene, the proposed method can realize high quality SAR imaging for non-sparse scene. Compared to the existing method, the proposed method has better reconstruction quality for region containing distinct edges and lines, such as city and river. Both the simulation scene and real scene experiments demonstrate the effectiveness of the proposed method.
- synthetic aperture radar /
- compressed sensing /
- sparse representation /
- approximate observation /
- curvelet transform

HTML全文

图 1 条带式SAR成像几何示意图

Figure 1. Geometry of strip mode SAR

下载: 全尺寸图片幻灯片

图 2 MSR-CS-SAR算法的流程

Figure 2. Flow block of MSR-CS-SAR algorithm

下载: 全尺寸图片幻灯片

图 3 SAR图像分解实验

Figure 3. Decomposition of SAR

下载: 全尺寸图片幻灯片

图 4 仿真场景示意图

Figure 4. Simulation scene

下载: 全尺寸图片幻灯片

图 5 36%降采样率仿真场景成像效果对比

Figure 5. Comparison of simulation scene SAR images under 36% downsample

下载: 全尺寸图片幻灯片

图 6 64%降采样率仿真场景成像效果对比

Figure 6. Comparison of simulation scene SAR images under 64% downsample

下载: 全尺寸图片幻灯片

图 7 81%降采样率仿真场景成像效果对比

Figure 7. Comparison of simulation scene SAR images under 81% downsample

下载: 全尺寸图片幻灯片

图 8 全采样率实测场景成像效果对比

Figure 8. Comparison of real scene SAR images under 100% downsample

下载: 全尺寸图片幻灯片

图 9 36%降采样率实测场景成像效果对比

Figure 9. Comparison of real scene SAR images under 36% downsample

下载: 全尺寸图片幻灯片

图 10 64%降采样率实测场景成像效果对比

Figure 10. Comparison of real scene SAR images under 64% downsample

下载: 全尺寸图片幻灯片

图 11 81%降采样率实测场景成像效果对比

Figure 11. Comparison of real scene SAR images under 81% downsample

下载: 全尺寸图片幻灯片

表 1 不同采样率下仿真场景成像结果的RMSE、PSNR、MSSIM

Table 1. RMSE, PSNR, MSSIM of simulation scene SAR images under different downsampling ratio

方法	36%降采样率			64%降采样率			81%降采样率
方法	RMSE	PSNR	MSSIM	RMSE	PSNR	MSSIM	RMSE	PSNR	MSSIM
基本方法	0.786	-9.268	0.053	0.735	-8.695	0.066	0.503	-5.390	0.109
文献[18]方法	0.316	-1.357	0.181	0.259	0.385	0.219	0.138	5.835	0.365
MSR-CS-SAR方法	0.309	-1.176	0.234	0.206	2.338	0.276	0.111	7.679	0.385

下载: 导出CSV

表 2 不同采样率下实测场景成像结果的RMSE、PSNR、MSSIM

Table 2. RMSE, PSNR, MSSIM of real scene SAR images under different downsampling ratio

方法	36%降采样率			64%降采样率			81%降采样率
方法	RMSE	PSNR	MSSIM	RMSE	PSNR	MSSIM	RMSE	PSNR	MSSIM
基本方法	3.387	7.771	0.142	4.157	5.993	0.173	4.831	4.688	0.174
文献[18]方法	0.180	33.28	0.951	0.099	38.44	0.975	0.093	39.01	0.979
MSR-CS-SAR方法	0.152	34.71	0.950	0.094	38.83	0.978	0.077	40.65	0.985

下载: 导出CSV

参考文献(21)

[1]	吴一戎, 洪文, 张冰尘, 等. 稀疏微波成像研究进展[J]. 雷达学报, 2014, 3(4): 383-395. https://www.cnki.com.cn/Article/CJFDTOTAL-LDAX201404002.htm WU Y R, HONG W, ZHANG B C, et al. Current developments of sparse microwave imaging[J]. Journal of Radars, 2014, 3(4): 383-395(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-LDAX201404002.htm
[2]	倪嘉成, 张群, 顾福飞, 等. 基于马尔科夫链的单站SAR海面场景宽幅高分成像算法[J]. 航空学报, 2016, 37(12): 3793-3802. https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201612024.htm NI J C, ZHANG Q, GU F F, et al. Mono-static SAR HRWS imaging algorithm of sea surface based Markov chain[J]. Acta Aeronautica et Astronautica Sinica, 2016, 37(12): 3793-3802(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201612024.htm
[3]	DONOHO D. Compressed sensing[J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306. doi: 10.1109/TIT.2006.871582
[4]	CANDES E J, WAKIN M B. An introduction to compressive sampling[J]. IEEE Signal Processing Magazine, 2008, 25(2): 21-30. doi: 10.1109/MSP.2007.914731
[5]	SAMADI S, CETIN M, MASNADI-SHIRAZI M A. Sparse representation-based synthetic aperture radar imaging[J]. IET Radar Sonar and Navigation, 2009, 5(2): 182-193.
[6]	FANG J, XU Z B, ZHANG B C, 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
[7]	顾福飞, 张群, 杨秋, 等. 基于NCS算子的大斜视SAR压缩感知成像方法[J]. 雷达学报, 2016, 5(1): 16-24. https://www.cnki.com.cn/Article/CJFDTOTAL-LDAX201601004.htm GU F F, ZHANG Q, YANG Q, et al. Compressed sensing imaging algorithm for high-squint SAR based on NCS operator[J]. Journal of Radars, 2016, 5(1): 16-24(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-LDAX201601004.htm
[8]	BI H, BI G A, ZHANG B C, et al. From theory to application: Realtime sparse SAR imaging[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(4): 2928-2936. doi: 10.1109/TGRS.2019.2958067
[9]	SHKVARKO Y V, Y ANEZ J I, AMAO J A. Radar/SAR image resolution enhancement via unifying descriptive experiment design regularization and wavelet-domain processing[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(2): 152-156. doi: 10.1109/LGRS.2015.2502539
[10]	ZHANG C L, LIU Y L, WAN F Y, et al. Multi-faults diagnosis of rolling bearings via adaptive customization of flexible analytical wavelet bases[J]. Chinese Journal of Aeronautics, 2020, 33(2): 407-417. doi: 10.1016/j.cja.2019.03.014
[11]	彭才, 常智, 朱仕军. 基于曲波变换的地震数据去噪方法[J]. 石油勘探, 2008, 47(5): 461-464. https://www.cnki.com.cn/Article/CJFDTOTAL-SYWT200805008.htm PENG C, CHANG Z, ZHU S J. Noise elimination method based on curvelet trans-form[J]. Geophysical Prospecting for Petroleum, 2008, 47(5): 461-464(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-SYWT200805008.htm
[12]	CETIN M, STOJANOVIC I, ONHON O, et al. Sparsity-driven synthetic aperture radar imaging: Reconstruction, autofocusing, moving targets and compressed sensing[J]. IEEE Signal Processing Magazine, 2014, 31(4): 27-40. doi: 10.1109/MSP.2014.2312834
[13]	SADEGH S, MUJDAT C, MOHAMMAD A M. Multiple feature enhanced SAR imaging using sparsity in combined dictionaries[J]. IEEE Geoscience and Remote Sensing Letters, 2013, 10(4): 821-825. doi: 10.1109/LGRS.2012.2225016
[14]	SHEN F F, ZHAO G H, LIU Z C, et al. SAR imaging with structural sparse representation[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(8): 3902-3910. doi: 10.1109/JSTARS.2014.2364294
[15]	MICHAL A, MICHAEL E, ALFRED B. K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation[J]. IEEE Transactions on Signal Processing, 2006, 54(11): 4311-4322. doi: 10.1109/TSP.2006.881199
[16]	胡长雨, 汪玲, 朱栋强. 结合字典学习技术的ISAR稀疏成像方法[J]. 电子与信息学报, 2019, 41(7): 1735-1742. https://www.cnki.com.cn/Article/CJFDTOTAL-DZYX201907028.htm HU C Y, WANG L, ZHU D Q. Sparse ISAR imaging exploiting dictionary learning[J]. Journal of Electronics & Information Technology, 2019, 41(7): 1735-1742(in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-DZYX201907028.htm
[17]	LI F, XIN L, GUO Y, et al. A framework of mixed sparse representations for remote sensing images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(2): 1210-1221.
[18]	LI B, LIU F L, ZHOU C B, et al. Mixed sparse representation for approximated observation-based compressed sensing radar imaging[J]. Journal of Applied Remote Sensing, 2018, 12(3): 1-21.
[19]	DABECHIES I, DEFRISE M, DEMOL C. An iterative thresholding algorithm for linear inverse problems with a sparsity constraint[J]. Communications on Pure and Applied Mathmatics, 2004, 57(11): 1413-1457.
[20]	RANEY R K, RUNGE H, BAMLER R, et al. Precision SAR processing using chirp scaling[J]. IEEE Transactions on Geoscience and Remote Sensing, 1994, 32(4): 786-799.
[21]	CANDES E, DEMANET L, DONOHO D. Fast discrete curvelet transforms[J]. Multiscale Modeling & Simulation, 2006, 5(3): 861-899.