Deep Learning-Based SAR Target Recognition: A Dual-Perspective Survey of Closed Set and Open Set
Source abstract
Abstract
Owing to the all-weather, day-and-night imaging capability of Synthetic Aperture Radar (SAR), SAR automatic target recognition (ATR) has long been a central focus in academia and industry. Since 2013, deep learning has become the dominant paradigm for SAR ATR owing to its end-to-end learning capability and robust feature-extraction capacity. To the best of our knowledge, this work provides the first systematic survey of SAR target recognition from dual closed-set and open-set perspectives and identifies four major performance bottlenecks: data scarcity, algorithmic limitations, hardware constraints, and application barriers. To address the first three bottlenecks, an in-depth analysis of closed-set solutions is presented, covering data augmentation, network optimization, and lightweight architectures. For the fourth challenge, a comprehensive analysis of open-set SAR recognition methods is provided. The intrinsic relationship and distinctions between closed-set and open-set recognition are further examined. To tackle the open-set challenge, an enhanced domain-adaptive algorithm for open-set recognition is proposed. Experiments on the OpenSAR and FUSAR datasets demonstrate at least a 3% improvement in open-set accuracy (OSA) over seven recent domain-adaptation algorithms. The rejection rate of unknown targets (RRU) reaches 80.30%, demonstrating a strong ability to distinguish unknown-class targets and offering practical insights for future research. Finally, potential directions for advancing SAR ATR are outlined, providing a comprehensive reference for the continued development of deep-learning-based SAR recognition.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced SAR Imaging Techniques · Synthetic Aperture Radar (SAR) Applications and Techniques · Domain Adaptation and Few-Shot Learning
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Automation Department, School of Information Science and Engineering, East China University of Science and Technology, Shanghai 200237, China
References (100+)
Show source-provided references
1. SAR ATR performance using a conditionally Gaussian model10.1109/7.913670 ↗
2. SAR Automatic Target Recognition Using Discriminative Graphical Models10.1109/TAES.2013.120340 ↗
3. Meth, R., and Chellappa, R. (1999, January 15–19). Feature matching and target recognition in synthetic aperture radar imagery. Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP99), Phoenix, AZ, USA.10.1109/ICASSP.1999.757555 ↗
4. Support vector machines for SAR automatic target recognition10.1109/7.937475 ↗
5. Ultralytics (2025, November 14). YOLOv8: The Next-Generation YOLO Model. GitHub, 2023. Available online: https://github.com/ultralytics/ultralytics.
6. Zheng, Y., Shcherbakova, G., Rusyn, B., Sachenko, A., Volkova, N., Kliushnikov, I., and Antoshchuk, S. (2025). Wavelet Transform Cluster Analysis of UAV Images for Sustainable Development of Smart Regions Due to Inspecting Transport Infrastructure. Sustainability, 17.10.3390/su17030927 ↗
7. Liao, Z., Zhao, Y., Shan, X., Yan, Y., Liu, C., Lu, L., Ji, X., and Chen, J. (2025). RT-DETRv4: Painlessly Furthering Real-Time Object Detection with Vision Foundation Models. arXiv.
8. New Discrimination Features for SAR Automatic Target Recognition10.1109/LGRS.2012.2210385 ↗
9. Chen, S., and Wang, H. (November, January 30). SAR target recognition based on deep learning. Proceedings of the 2014 International Conference on Data Science and Advanced Analytics (DSAA), Shanghai, China.10.1109/DSAA.2014.7058124 ↗
10. Classification on the Monogenic Scale Space: Application to Target Recognition in SAR Image10.1109/TIP.2015.2421440 ↗
11. SARATR-X: Toward Building a Foundation Model for SAR Target Recognition10.1109/TIP.2025.3531988 ↗
12. Few-Shot Class-Incremental SAR Target Recognition Based on Hierarchical Embedding and Incremental Evolutionary Network
13. SAR ATR by a combination of convolutional neural network and support vector machines10.1109/TAES.2016.160061 ↗
14. Target Classification Using the Deep Convolutional Networks for SAR Images10.1109/TGRS.2016.2551720 ↗
15. Synthetic Aperture Radar Image Synthesis by Using Generative Adversarial Nets10.1109/LGRS.2017.2699196 ↗
16. SAR Automatic Target Recognition Based on Euclidean Distance Restricted Autoencoder10.1109/JSTARS.2017.2670083 ↗
17. Multiple Feature Aggregation Using Convolutional Neural Networks for SAR Image-Based Automatic Target Recognition10.1109/LGRS.2018.2865608 ↗
18. SAR Targets Classification Based on Deep Memory Convolution Neural Networks and Transfer Parameters10.1109/JSTARS.2018.2836909 ↗
19. Semi-Supervised SAR ATR via Multi-Discriminator Generative Adversarial Network10.1109/JSEN.2019.2915379 ↗
20. SAR Target Image Classification Based on Transfer Learning and Model Compression10.1109/LGRS.2018.2876378 ↗
21. LDGAN: A Synthetic Aperture Radar Image Generation Method for Automatic Target Recognition10.1109/TGRS.2019.2957453 ↗
22. What, Where, and How to Transfer in SAR Target Recognition Based on Deep CNNs10.1109/TGRS.2019.2947634 ↗
23. FEC: A Feature Fusion Framework for SAR Target Recognition Based on Electromagnetic Scattering Features and Deep CNN Features10.1109/TGRS.2020.3003264 ↗
24. SAR Target Classification Based on Integration of ASC Parts Model and Deep Learning Algorithm10.1109/JSTARS.2021.3116979 ↗
25. SAR Target Classification Using the Multikernel-Size Feature Fusion-Based Convolutional Neural Network10.1109/TGRS.2021.3106915 ↗
26. Few-Shot SAR Target Classification via Metalearning10.1109/TGRS.2021.3058249 ↗
27. ADCG: A Cross-Modality Domain Transfer Learning Method for Synthetic Aperture Radar in Ship Automatic Target Recognition10.1109/TGRS.2023.3313204 ↗
28. HDSS-Net: A Novel Hierarchically Designed Network With Spherical Space Classifier for Ship Recognition in SAR Images10.1109/TGRS.2023.3332137 ↗
29. MGSFA-Net: Multiscale Global Scattering Feature Association Network for SAR Ship Target Recognition10.1109/JSTARS.2024.3357171 ↗
30. SAR Target Incremental Recognition Based on Features with Strong Separability
31. Li, Y.-B., Zhou, C., and Wang, N. (2010, January 22–24). A survey on feature extraction of SAR Images. Proceedings of the 2010 International Conference on Computer Application and System Modeling (ICCASM 2010), Taiyuan, China.10.1109/ICCASM.2010.5619316 ↗
32. Automatic Target Recognition in Synthetic Aperture Radar Imagery: A State-of-the-Art Review10.1109/ACCESS.2016.2611492 ↗
33. Study on Vessel Classification in SAR Imagery: A Survey
34. Recent Advances in Deep-Learning-Based SAR Image Target Detection and Recognition10.1109/JSTARS.2025.3543531 ↗
35. Review of Synthetic Aperture Radar Automatic Target Recognition: A Dual Perspective on Classical and Deep Learning Techniques10.1109/JSTARS.2025.3589804 ↗
36. Automatic Target Recognition on Synthetic Aperture Radar Imagery: A Survey10.1109/MAES.2021.3049857 ↗
37. Transfer Adaptation Learning for Target Recognition in SAR Images: A Survey10.1109/JSTARS.2024.3434448 ↗
38. Open World Object Detection: A Survey10.1109/TCSVT.2024.3480691 ↗
39. Keydel, E.R., Lee, S.W., and Moore, J.T. (1996, January 10). MSTAR extended operating conditions: A tutorial. Proceedings of the Aerospace/Defense Sensing and Controls, Orlando, FL, USA.
40. Sandia National Laboratories (2025, January 12). SAR Data, Pathfinder Radar ISR & SAR Systems, Available online: https://www.sandia.gov/radar/pathfinder-radar-isr-and-syntheticaperture-radar-sar-systems/complex-data/.
41. Improving a template-based classifier in a SAR automatic target recognition system by using 3-D target information
42. Kusk, A., Abulaitijiang, A., and Dall, J. (2016, January 6–9). Synthetic SAR image generation using sensor, terrain and target models. Proceedings of the EUSAR 2016: 11th European Conference on Synthetic Aperture Radar, Hamburg, Germany.
43. Zhang, T., Zhang, X., Ke, X., Zhan, X., Shi, J., Wei, S., Pan, D., Li, J., Su, H., and Zhou, Y. (2020). LS-SSDD-v1.0: A Deep Learning Dataset Dedicated to Small Ship Detection from Large-Scale Sentinel-1 SAR Images. Remote Sens., 12.10.3390/rs12182997 ↗
44. AIR-PolSAR-Seg-2.0: Polarimetric SAR ground terrain classification dataset for large-scale complex scenes
45. SAR-AIRcraft-1.0: High-resolution SAR aircraft detection and recognition dataset
46. FAIR-CSAR: A Benchmark Dataset for Fine-Grained Object Detection and Recognition Based on Single-Look Complex SAR Images
47. Hu, Y., Li, Y., and Pan, Z. (2021). A dual-polarimetric SAR ship detection dataset and a memory-augmented autoencoder-based detection method. Sensors, 21.10.3390/s21248478 ↗
48. OpenSARUrban: A Sentinel-1 SAR Image Dataset for Urban Interpretation10.1109/JSTARS.2019.2954850 ↗
49. Franceschetti, G., Migliaccio, M., and Riccio, D. (1995, January 10–14). The SAR simulation: An overview. Proceedings of the 1995 International Geoscience and Remote Sensing Symposium (IGARSS’95), Firenze, Italy.
50. A canonical problem in electromagnetic backscattering from buildings10.1109/TGRS.2002.802459 ↗
51. SAR raw signal simulation for urban structures10.1109/TGRS.2003.814626 ↗
52. Radar Imaging Simulation for Urban Structures10.1109/LGRS.2010.2051214 ↗
53. Hammer, H., and Schulz, K. (2009, January 29). Coherent simulation of SAR images. Proceedings of the Image and Signal Processing for Remote Sensing XV; 74771G (2009), Berlin, Germany.10.1117/12.830380 ↗
54. Ghost Persistent Scatterers Related to Multiple Signal Reflections10.1109/LGRS.2011.2134066 ↗
55. Auer, S., Gernhardt, S., and Bamler, R. (2011, January 24–29). Ghost persistent scatterers related to signal reflections between adjacent buildings. Proceedings of the 2011 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Vancouver, BC, Canada.10.1109/IGARSS.2011.6049348 ↗
56. Niu, S., Qiu, X., Peng, L., and Lei, B. (July, January 6). Parameter Prediction Method of SAR Target Simulation Based on Convolutional Neural Networks. Proceedings of the EUSAR 2018: 12th European Conference on Synthetic Aperture Radar, Aachen, Germany.
57. Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. (2014). Generative Adversarial Nets, MIT Press. Available online: https://arxiv.org/pdf/1406.2661.
58. Schwegmann, C.P., Kleynhans, W., Salmon, B.P., Mdakane, L.W., and Meyer, R.G.V. (2017, January 23–28). Synthetic aperture radar ship discrimination, generation and latent variable extraction using information maximizing generative adversarial networks. Proceedings of the 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Fort Worth, TX, USA.10.1109/IGARSS.2017.8127440 ↗
59. Image Data Augmentation for SAR Sensor via Generative Adversarial Nets10.1109/ACCESS.2019.2907728 ↗
60. Mroueh, Y., Sercu, T., and Goel, V. (2017). McGan: Mean and Covariance Feature Matching GAN. arXiv.
61. Liu, L., Pan, Z., Qiu, X., and Peng, L. (2018, January 22–27). SAR Target Classification with CycleGAN Transferred Simulated Samples. Proceedings of the IGARSS 2018-2018 IEEE International Geoscience and Remote Sensing Symposium, Valencia, Spain.10.1109/IGARSS.2018.8517866 ↗
62. SAR Target Recognition Based on Cross-Domain and Cross-Task Transfer Learning10.1109/ACCESS.2019.2948618 ↗
63. Chen, Z., Dai, M., Cong, Y., and Lu, R. (June, January 31). Attribute-Controlled SAR Target Image Generation Driven by Diffusion Models. Proceedings of the 2024 5th International Conference on Electronic Communication and Artificial Intelligence (ICECAI), Shenzhen, China.10.1109/ICECAI62591.2024.10675205 ↗
64. DiffuSAR: Frequency Domain-Aware Diffusion Model for SAR Image Generation10.1109/JSTARS.2025.3563798 ↗
65. Zhang, H., and Wu, G. (2025, January 20–22). DiT-DDIM: Invisible Diffusion Image Generation Model Based on Transformer Framework. Proceedings of the 2025 IEEE International Conference on Pattern Recognition, Machine Vision and Artificial Intelligence (PRMVAI), Loudi, China.10.1109/PRMVAI65741.2025.11108570 ↗
66. Crump, T., Kompella, S., Kam, C., and Tassiulas, L. (November, January 28). SAR Data Generation Using Denoising Diffusion Probabilistic Models. Proceedings of the MILCOM 2024—2024 IEEE Military Communications Conference (MILCOM), Washington, DC, USA.10.1109/MILCOM61039.2024.10773802 ↗
67. Xie, G., Lu, H., Zhang, X., and He, K. (2025, January 6–11). Advancing SAR Image Robustness: Integrating Diffusion Models for Adversarial Purification and Speckle Noise Suppression. Proceedings of the ICASSP 2025–2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Hyderabad, India.10.1109/ICASSP49660.2025.10888393 ↗
68. DADSR: Degradation-Aware Diffusion Super-Resolution Model for Object-Level SAR Image10.1109/JSTARS.2025.3590421 ↗
69. SAR Despeckling Using a Denoising Diffusion Probabilistic Model10.1109/LGRS.2023.3270799 ↗
70. Complex-valued Conditional Diffusion Model with Physical Scattering Information for SAR Image Target Recognition10.1109/JSTARS.2025.3612376 ↗
71. SAR Image Target Recognition Using Diffusion Model and Scattering Information10.1109/LGRS.2024.3466233 ↗
72. Zhou, Y., Li, Y., Xie, W., and Li, L. (2021). A Convolutional Neural Network Combined with Attributed Scattering Centers for SAR ATR. Remote Sens., 13.10.3390/rs13245121 ↗
73. Integrating the Reconstructed Scattering Center Feature Maps with Deep CNN Feature Maps for Automatic SAR Target Recognition
74. Multiscale CNN Based on Component Analysis for SAR ATR
75. A Novel SAR Target Recognition Method Combining Electromagnetic Scattering Information and GCN
76. Multilevel Scattering Center and Deep Feature Fusion Learning Framework for SAR Target Recognition
77. Pan, Y., Tang, L., Jing, D., Tang, W., and Zhou, S. (2019, January 11–13). Efficient and Lightweight Target Recognition for High Resolution Spaceborne SAR Images. Proceedings of the 2019 IEEE International Conference on Signal, Information and Data Processing (ICSIDP), Chongqing, China.10.1109/ICSIDP47821.2019.9173445 ↗
78. Imbalanced High-Resolution SAR Ship Recognition Method Based on a Lightweight CNN
79. Lightweight Two-Stream Convolutional Neural Network for SAR Target Recognition10.1109/LGRS.2020.2983718 ↗
80. Wang, J., Jiang, T., Cui, Z., Cao, Z., and Cao, C. (2022, January 17–22). A Knowledge Distillation Method Based on IQE Attention Mechanism for Target Recognition in SAR Imagery. Proceedings of the IGARSS 2022—2022 IEEE International Geoscience and Remote Sensing Symposium, Kuala Lumpur, Malaysia.10.1109/IGARSS46834.2022.9883376 ↗
81. Slim and Efficient Neural Network Design for Resource-Constrained SAR Target Recognition10.3390/rs10101618 ↗
82. Open Set Incremental Learning for Automatic Target Recognition10.1109/TGRS.2019.2891266 ↗
83. Dang, S., Cao, Z., Cui, Z., and Pi, Y. (2019, January 26–29). Open Set SAR Target Recognition Using Class Boundary Extracting. Proceedings of the 2019 6th Asia-Pacific Conference on Synthetic Aperture Radar (APSAR), Xiamen, China.10.1109/APSAR46974.2019.9048316 ↗
84. Training SAR-ATR Models for Reliable Operation in Open-World Environments10.1109/JSTARS.2021.3068944 ↗
85. Zeng, Z., Sun, J., Xu, C., and Wang, H. (2021). Unknown SAR Target Identification Method Based on Feature Extraction Network and KLD-RPA Joint Discrimination. Remote Sens., 13.10.3390/rs13152901 ↗
86. Lu, Y., Pei, J., Zhang, Y., Huo, W., Huang, Y., Wang, W., and Yang, J. (2024, January 7–12). A SAR Open-Set Recognition Method Aided by Hierarchically Reconstructive Latent Representation Learning. Proceedings of the IGARSS 2024—2024 IEEE International Geoscience and Remote Sensing Symposium, Athens, Greece.10.1109/IGARSS53475.2024.10642884 ↗
87. An Entropy-Awareness Meta-Learning Method for SAR Open-Set ATR
88. Lu, Y., Pei, J., Zhang, Y., Huang, Y., Yang, J., and Wu, J. (2025). CG2OSR: A Class Gaussian Guided Framework for SAR Open-Set Target Recognition. IEEE Trans. Aerosp. Electron. Syst., 1–19.10.1109/TAES.2025.3618812 ↗
89. An Open Set Recognition Method for SAR Targets Based on Multitask Learning10.1109/LGRS.2021.3079418 ↗
90. Geng, Z., Li, W., Yu, X., Zhu, D.-Y., and Zhang, G. (2023, January 24–26). Out-of-Library SAR Target Recognition with Deep Learning from Synthetic Data and Multiview Information Fusion. Proceedings of the 2023 24th International Radar Symposium (IRS), Berlin, Germany.10.23919/IRS57608.2023.10172440 ↗
91. Li, W., Zhou, H., Geng, Z., and Zhu, D. (2024, January 12–14). Monostatic and Bistatic SAR Automatic Target Recognition with Multi-Aspect Feature Fusion in Open Environment. Proceedings of the 2024 9th International Conference on Signal and Image Processing (ICSIP), Nanjing, China.10.1109/ICSIP61881.2024.10671486 ↗
92. Xiao, X., Li, Z., Wu, X., and Wang, H. (2024, January 7–12). Embedding Attribute Scattering Center with Convolutional Prototype Learning for SAR Open-Set Recognition. Proceedings of the IGARSS 2024—2024 IEEE International Geoscience and Remote Sensing Symposium, Athens, Greece.10.1109/IGARSS53475.2024.10642410 ↗
93. Reciprocal Point Learning Network with Large Electromagnetic Kernel for SAR Open-Set Recognition10.1109/TAES.2025.3572447 ↗
94. Li, Y., Ren, H., Yu, X., Zhang, C., Zou, L., and Zhou, Y. (2023, January 24–27). Threshold-Free Adaptive Open-Set Classifier for SAR Target Recognition. Proceedings of the IET International Radar Conference (IRC 2023), Chongqing, China.10.1049/icp.2024.1286 ↗
95. Threshold-Free Open-Set Learning Network for SAR Automatic Target Recognition10.1109/JSEN.2024.3354966 ↗
96. GANFlow: A Hybrid Model for SAR Image Target Open-Set Recognition Based on GAN and the Flow-Based Module10.1109/JSTARS.2025.3542738 ↗
97. Open-Set Domain Adaptation via Adaptive Separation and Two-Level Alignment with Application in SAR Target Recognition10.1109/JSTARS.2025.3615206 ↗
98. Open Set Recognition with Incremental Learning for SAR Target Classification10.1109/TGRS.2023.3283423 ↗
99. Open-Set SAR Target Recognition Based on Observation-Invariant Contrastive Learning10.1109/JSEN.2025.3603159 ↗
100. Combining Local Electromagnetic Scattering and Global Structure Features for SAR Open Set Recognition10.1109/JSTARS.2025.3568330 ↗