SAR-ATR-methods

This repository contains the supplementary file for article "Review of Synthetic Aperture Radar Automatic Target Recognition: A Dual Perspective on Classical and Deep Learning Techniques"

├── LICENSE                 
├── README.md
└── SAR_ATR_methods_Table.csv       # paper review database

Taxonomy of the Detection Methods:

1. Traditional Methods

A. Template-Based Methods

B. Model-Based Methods

C. CFAR-Based Methods

• One-parameter CFAR
• Two-parameter CFAR
• Clutter Modeling and Applications

D. Machine Learning Methods

2. Modern Deep Learning Methods

A. Architecture-Based Classification

• Non-CNN Deep Models
• CNN-Based Methods
  • Standard CNNs (Borrowed from Computer Vision)
  • Shallow and deep CNNs
  • Two-Stage and One-Stage Detectors
  • Attention-based CNNs
  • Capsule Networks
• Hybrid Architectures

B. Learning Paradigm Classification

• Supervised Learning
• Semi-supervised Learning
• Unsupervised Learning
• Few-shot Learning
• Transfer Learning
• Adversarial Attack
• Data Augmentation Strategies
  • Classical data augmentation
  • Simulation-based generation
  • GAN-based new samples
• Scattering-Based Feature Enhancement
  • Attributed Scattering Center Based
  • Scattering topology
• Multi-modal Data Processing
  • Polarimetric SAR
  • SAR-optical fusion
  • Complex data
  • Multi-resolution SAR

C. Operational and Task-Specific Methods

• Real-time Detection Systems
• Robust Recognition Systems
• Small-Target Detection
• Multitask

DATASETS IN SAR OBJECT RECOGNITION

Dataset	Year	Sensor/Source	# Img	Img size	Res [m]	Band	Pol	Target
MSTAR [1]	1995	Aerial X-band radar	14,577	128×128	0.3	X-band	Single	Military vehicles
Sandia MiniSAR [2]	2006	MiniSAR	3,927	224×224	0.1	Ku-band	Single, Dual	Terrestrial targets in urban areas
SARSim [3]	2016	simulated data	21,168	139×139	0.1 to 0.3	X-band	Single	Military vehicles
OpenSARShip-1.0 [4]	2017	Sentinel-1	11,346	Various	2.7×22 to 3.5×22 and 20×22	C-band	Single, Dual	Ships (AIS types)
OpenSARShip-2.0 [5]	2017	Sentinel-1	34,528	Various	2.7×22 to 3.5×22 and 20×22	C-band	Single, Dual	Ships (AIS types)
SSDD [6]	2017	RADARSAT-2, TerraSAR-X, Sentinel-1	1,16	Various 214 to 668	1 to 15	C-band X-band	Singl, Dual, Quad	Ships
SAMPLE [7]	2019	MSTAR + simulated data	2,732	128×128	0.3	X-band	Single	Military vehicles
SAR-Ship-Dataset [8]	2019	Gaofen-3, Sentinel-1	43,819	256×256	1.7×4.3 to 25	C-band	Single, Dual	Ships
AIR-SARShip-1.0 [9]	2019	Gaofen-3	31	3000×3000 to 4140×4140	1 to 3	C-band	Single	Ships
AIR-SARShip-2.0 [10]	2020	Gaofen-3	300	~1000×1000	1 to 3	C-band	Single	Ships
HRSID [11]	2020	Sentinel-1, TerraSAR-X, TanDEM-X	5,604	800×800	0.5 to 3	C-band X-band	Dual, Quad	Ships
LS-SSDD-v1.0 [12]	2020	Sentinel-1	9	800×800	5x20	C-band	Single, Dual	Small ships
FUSAR-Ship Dataset v1.0 [13]	2020	Gaofen-3	16,144	512×512	1.12 × 1.73	C-band	Single, Dual	Ships (AIS messages)
DSSDD [14]	2021	TerraSAR-X, Sentinel-1, Gaofen-3	1,236	256×256	Not specified	C-band X-band	Single, Dual	Ships
SRSDD-v1.0 [15]	2021	Gaofen-3	666	1024×1024	1	C-band	Single	Ships
xView3-SAR [16]	2022	Sentinel-1	991	~29,400× 24,400	~20	C-band	Single, Dual	Ships, offshore structures
SADD [17]	2022	TerraSAR-X	2,966	224×224	0.5 to 3	X-band	Single	Aircraft
MSAR [18]	2022	Multiple	28,499	256×256 to 2048×2048	1	C-band	Quad	Terrestrial and maritime targets
SAR-AIRcraft [19]	2023	Gaofen-3	4,368	800×800 to 1500×1500	1	C-band	Single	Aircraft
SIVED [20]	2023	simulated data	1,044	512×512	0.1 to 0.3	Ku-band Ka-band X-band	Single	Vehicles
MRSSD [21]	2023	Capella, ICEYE, TerraSAR-X, Paz, Alos PALSAR, Sentinel-1	11,59	512×512	0.5 to 100	L-band C-band X-band	Single, Dual, Quad	Ships
SARDet-100K [22]	2024	Gaofen-3, Sentinel-1, TanDEM-X, RADARSAT-2, Alos-PALSAR, Capella, ICEYE, Kompsat-5, RISAT-1	116,598	512×512	0.5m to 3m	L-band C-band X-band	Single, Dual	Ships, vehicles, others

References

• [1] E. R. Keydel, S. W. Lee, and J. T. Moore, ‘MSTAR extended operating conditions: a tutorial’, presented at the Aerospace/Defense Sensing and Controls, E. G. Zelnio and R. J. Douglass, Eds., Orlando, FL, Jun. 1996, pp. 228–242. doi: 10.1117/12.242059.
• [2] ‘SAR Data’, Pathfinder Radar ISR & SAR Systems. Accessed: Jan. 12, 2025. [Online]. Available: https://www.sandia.gov/radar/pathfinder-radar-isr-and-synthetic-aperture-radar-sar-systems/complex-data/
• [3] ‘Synthetic SAR Image Generation using Sensor, Terrain and Target Models’. Accessed: Jan. 12, 2025. [Online]. Available: https://ieeexplore-1ieee-1org-100001bck0bcc.han.wat.edu.pl/document/7559326
• [4] ‘OpenSARShip: A Dataset Dedicated to Sentinel-1 Ship Interpretation’. Accessed: Jan. 12, 2025. [Online]. Available: https://ieeexplore-1ieee-1org-100001bck0bdf.han.wat.edu.pl/document/8067489
• [5] ‘OpenSARShip 2.0: A large-volume dataset for deeper interpretation of ship targets in Sentinel-1 imagery’. Accessed: Jan. 12, 2025. [Online]. Available: https://ieeexplore-1ieee-1org-100001bck0bdf.han.wat.edu.pl/document/8124929
• [6] T. Zhang et al., ‘SAR Ship Detection Dataset (SSDD): Official Release and Comprehensive Data Analysis’, Remote Sensing, vol. 13, no. 18, p. 3690, Sep. 2021, doi: 10.3390/rs13183690.
• [7] B. Lewis, T. Scarnati, E. Sudkamp, J. Nehrbass, S. Rosencrantz, and E. Zelnio, ‘A SAR dataset for ATR development: the Synthetic and Measured Paired Labeled Experiment (SAMPLE)’, in Algorithms for Synthetic Aperture Radar Imagery XXVI, SPIE, May 2019, pp. 39–54. doi: 10.1117/12.2523460.
• [8] ‘A SAR Dataset of Ship Detection for Deep Learning under Complex Backgrounds’. Accessed: Jan. 12, 2025. [Online]. Available: https://www.mdpi.com/2072-4292/11/7/765
• [9] S. Xian et al., ‘AIR-SARShip-1.0: High-resolution SAR Ship Detection Dataset’, 2020. Accessed: Dec. 23, 2024. [Online]. Available: https://www.semanticscholar.org/paper/AIR-SARShip-1.0%3A-High-resolution-SAR-Ship-Detection-Xian-Wang/bdf2e09c65e857fd608df15ef5c317a9267489b3
• [10] ‘雷达学报’. Accessed: Jan. 12, 2025. [Online]. Available: https://radars.ac.cn/web/data/getData?newsColumnId=74fe223a-0b01-4830-8d99-1ba276e67ad8&pageType=en
• [11] S. Wei, X. Zeng, Q. Qu, M. Wang, H. Su, and J. Shi, ‘HRSID: A High-Resolution SAR Images Dataset for Ship Detection and Instance Segmentation’, IEEE Access, vol. 8, pp. 120234–120254, 2020, doi: 10.1109/ACCESS.2020.3005861.
• [12] T. Zhang et al., ‘LS-SSDD-v1.0: A Deep Learning Dataset Dedicated to Small Ship Detection from Large-Scale Sentinel-1 SAR Images’, Remote Sensing, vol. 12, no. 18, p. 2997, Sep. 2020, doi: 10.3390/rs12182997.
• [13] X. Hou, W. Ao, Q. Song, J. Lai, H. Wang, and F. Xu, ‘FUSAR-Ship: building a high-resolution SAR-AIS matchup dataset of Gaofen-3 for ship detection and recognition’, Sci. China Inf. Sci., vol. 63, no. 4, p. 140303, Mar. 2020, doi: 10.1007/s11432-019-2772-5.
• [14] Y. Hu, Y. Li, and Z. Pan, ‘A Dual-Polarimetric SAR Ship Detection Dataset and a Memory-Augmented Autoencoder-Based Detection Method’, Sensors (Basel), vol. 21, no. 24, p. 8478, Dec. 2021, doi: 10.3390/s21248478.
• [15] S. Lei, D. Lu, X. Qiu, and C. Ding, ‘SRSDD-v1.0: A High-Resolution SAR Rotation Ship Detection Dataset’, Remote Sensing, vol. 13, no. 24, p. 5104, Dec. 2021, doi: 10.3390/rs13245104.
• [16] F. Paolo et al., ‘xView3-SAR: Detecting Dark Fishing Activity Using Synthetic Aperture Radar Imagery’, Nov. 05, 2022, arXiv: arXiv:2206.00897. doi: 10.48550/arXiv.2206.00897.
• [17] P. Zhang et al., ‘SEFEPNet: Scale Expansion and Feature Enhancement Pyramid Network for SAR Aircraft Detection With Small Sample Dataset’, IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing, vol. 15, pp. 3365–3375, 2022, doi: 10.1109/JSTARS.2022.3169339.
• [18] ‘雷达学报’. Accessed: Jan. 12, 2025. [Online]. Available: https://radars.ac.cn/web/data/getData?dataType=MSAR
• [19] Y. KANG, Y. KANG, X. ZENG, Y. WANG, T. ZHANG, and X. SUN, ‘SAR-AIRcraft-1.0: High-resolution SAR Aircraft Detection and Recognition Dataset’, Journal of Radars, vol. 12(4), doi: 10.12000/JR23043.
• [20] ‘SIVED: A SAR Image Dataset for Vehicle Detection Based on Rotatable Bounding Box’. Accessed: Jan. 12, 2025. [Online]. Available: https://www.mdpi.com/2072-4292/15/11/2825
• [21] M. F. Humayun, F. A. Bhatti, and K. Khurshid, ‘iVision MRSSD: A comprehensive multi-resolution SAR ship detection dataset for state of the art satellite based maritime surveillance applications’, Data in Brief, vol. 50, p. 109505, Oct. 2023, doi: 10.1016/j.dib.2023.109505.
• [22] Y. Li et al., ‘SARDet-100K: Towards Open-Source Benchmark and ToolKit for Large-Scale SAR Object Detection’, Sep. 30, 2024, arXiv: arXiv:2403.06534. doi: 10.48550/arXiv.2403.06534.