A Generative Adversarial Network-Based Method for Intelligent Detection of Military Targets with Few Training Samples for Embedded Edge Devices
Main Article Content
Abstract
With the development of artificial intelligence and intelligent algorithms, new technical methods of automatic military target recognition in complex environments have been provided. However, limited availability of labeled samples, significant scale variations in targets and computing resource limitations on embedded devices all still limit detection accuracy and the ability to deploy detection in real-time. In this regard, a generative adversarial network (GAN) based few-shot military target detection is presented. To enlarge the training data set, class-conditional sample generation, multi-scale feature encoding, target-region attention constraints and generated-sample quality screening are employed, and the joint training of real and generated samples is used to enhance the generalization ability of the model. Based on this, depthwise separable convolution, channel pruning, structural re-parameterization, mixed-precision quantization, knowledge distillation, and operator fusion are introduced to realize lightweight network design and optimize the network on the edge. Experimental results show that the proposed method achieves a precision of 91.2%, a recall of 88.9%, and an F1-score of 90.0%. The mAP@0.5 and mAP@0.5:0.95 reach 92.7% and 61.8%, respectively, while the model contains only 4.1 M parameters and requires 7.6 G FLOPs. On the NVIDIA Jetson Orin NX platform, the inference speed reaches 106.4 FPS. These results demonstrate that the method effectively balances few-shot detection accuracy, robustness in complex environments, and real-time inference requirements on embedded edge devices.
Downloads
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
References
X. Zhuang, D. Li, Y. Wang, et al., “Military target detection method based on EfficientDet and generative adversarial network,” Eng. Appl. Artif. Intell., vol. 132, Art. no. 107896, 2024, doi: 10.1016/j.engappai.2024. 107896.
A. Berkol, “Exploring AI-based techniques for the generation of synthetic infrared images and their practical applications,” in Proc. 2025 9th Int. Symp. Innov. Approaches Smart Technol. (ISAS), 2025, pp. 1–5, doi: 10.1109/ISAS66241.2025.11101900.
S. Cheong, W. Jung, Y. S. Lim, et al., “Thermal-infrared remote-target detection system for maritime rescue using 3-D game-based data augmentation with GAN,” IEEE Trans. Geosci. Remote Sens., vol. 62, pp. 1–13, 2024, doi: 10.1109/TGRS.2024.3454983.
Y. Zhang, J. Zhu, and K. Cao, “Few-shot military target detection with text-to-image and vision-language models,” in Proc. 2024 7th Int. Conf. Image Graph. Process., 2024, pp. 47–53, doi: 10.1145/3647649.3647657.
T. Zhuang, X. Liang, B. Xue, et al., “An in-vehicle real-time infrared object detection system based on deep learning with resource-constrained hardware,” Intell. Robot., vol. 4, no. 3, pp. 276–292, 2024, doi: 10.20517/i r.2024.18.
N. Jiang, W. Zhao, H. Wang, et al., “Lightweight super-resolution generative adversarial network for SAR images,” Remote Sens., vol. 16, no. 10, Art. no. 1788, 2024, doi: 10.3390/rs16101788.
W. Kong, H. Yang, M. Jia, et al., “CDBA-GAN: A conditional dual-branch attention generative adversarial network for robust sonar image generation,” Appl. Sci., vol. 15, no. 13, Art. no. 7212, 2025, doi: 10.3390/app15137212.
D. Gao, X. Wu, Z. Wen, et al., “Few-shot SAR vehicle target augmentation based on generative adversarial networks,” ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., vol. 10, pp. 83–90, 2024, doi: 10.5194/isprs-annals-X-1-2024-83-2024.
M. S. Alexiou and J. S. Mertoguno, “A survey on recent advancements in lightweight generative adversarial networks, their applications and datasets,” in Proc. 2023 IEEE 35th Int. Conf. Tools Artif. Intell. (ICTAI), 2023, pp. 269–278, doi: 10.1109/ICTAI59109.2023.00047.
Y. R. Musunuri, C. Kim, O. S. Kwon, et al., “Object detection using ESRGAN with a sequential transfer learning on remote sensing embedded systems,” IEEE Access, vol. 12, pp. 102313–102327, 2024, doi: 10.1109/ACCESS.2024.3432532.
X. Yi, H. Pan, H. Zhao, et al., “Cycle generative adversarial network based on gradient normalization for infrared image generation,” Appl. Sci., vol. 13, no. 1, Art. no. 635, 2023, doi: 10.3390/app13010635.
Z. Wang, C. Wang, Y. Chen, et al., “Target detection algorithm based on super-resolution color remote sensing image reconstruction,” J. Meas. Eng., vol. 12, no. 1, pp. 83–98, 2024, doi: 10.21595/jme.2023.23510.
S. Cheng, Y. Han, Z. Wang, et al., “An underwater object recognition system based on improved YOLOv11,” Electronics, vol. 14, no. 1, Art. no. 201, 2025, doi: 10.3390/electronics14010201.
T. H. Vu, S. K. Jagatheesaperumal, M. D. Nguyen, et al., “Applications of generative AI (GAI) for mobile and wireless networking: A survey,” IEEE Internet Things J., vol. 12, no. 2, pp. 1266–1290, 2024, doi: 10.1109/JIOT.2024.3487627.
A. Prabagaran, S. Kalavathi, N. M. Balamurugan, et al., “Secured heterogeneous data fusion using adversarial autoencoders in internet of military things,” in Proc. 2025 Int. Conf. Data Sci. Bus. Syst. (ICDSBS), 2025, pp. 1–6, doi: 10.1109/ICDSBS63635.2025.11031555.
M. Chemmakha, O. Habibi, and M. Lazaar, “Towards a deep learning approach for IoT attack detection based on a new generative adversarial network architecture and gated recurrent unit,” J. Netw. Syst. Manage., vol. 32, no. 4, Art. no. 96, 2024, doi: 10.1007/s10922-024-09873-1.
A. Guesmi, M. A. Hanif, B. Ouni, et al., “Physical adversarial attacks for camera-based smart systems: Current trends, categorization, applications, research challenges, and future outlook,” IEEE Access, vol. 11, pp. 109617–109668, 2023, doi: 10.1109/ACCESS.2023.3321118.
Z. Wang, B. Wang, C. Zhang, et al., “Robust feature-guided generative adversarial network for aerial image semantic segmentation against back-door attacks,” Remote Sens., vol. 15, no. 10, Art. no. 2580, 2023, doi: 10.3390/rs15102580.
C. Wang, S. Cheng, Y. Hou, et al., “Research on image damage assessment testing method based on deep learning,” in Proc. 2025 8th Int. Conf. Adv. Algorithms Control Eng. (ICAACE), 2025, pp. 1255–1260, doi: 10.1109/ICAACE65325.2025.11019570.
M. Andreoni, W. T. Lunardi, G. Lawton, et al., “Enhancing autonomous system security and resilience with generative AI: A comprehensive survey,” IEEE Access, vol. 12, pp. 109470–109493, 2024, doi: 10.1109/AC CESS.2024.3439363.
A. B. Rashid, A. K. Kausik, A. al Hassan Sunny, et al., “Artificial intelligence in the military: An overview of the capabilities, applications, and challenges,” Int. J. Intell. Syst., vol. 2023, no. 1, Art. no. 8676366, 2023, doi: 10.1155/2023/8676366.
M. A. Habib, M. A. H. Wadud, M. F. K. Patwary, et al., “Exploring progress in text-to-image synthesis: An in-depth survey on the evolution of generative adversarial networks,” IEEE Access, vol. 12, pp. 178401– 178440, 2024, doi: 10.1109/ACCESS.2024.3435541.
N. K. Son, A. K. Sangaiah, D. V. Medhane, et al., “Enhancing resilience in edge IoT devices against adversarial attacks,” IEEE Consum. Electron. Mag., vol. 14, no. 4, pp. 48–56, 2024, doi: 10.1109/MCE.2024.3522524.
C. Liu, X. Fu, Y. Wang, et al., “Overcoming data limitations: A few-shot specific emitter identification method using self-supervised learning and adversarial augmentation,” IEEE Trans. Inf. Forensics Security, vol. 19, pp. 500–513, 2023, doi: 10.1109/TIFS.2023.3324394.
X. Liu, G. Li, Z. Zhao, et al., “EAF-WGAN: Enhanced alignment fusion-Wasserstein generative adversarial network for turbulent image restoration,” IEEE Trans. Circuits Syst. Video Technol., vol. 33, no. 10, pp. 5605– 5616, 2023, doi: 10.1109/TCSVT.2023.3262685.
J. Yang, H. Chu, L. Guo, et al., “A weighted-transfer domain-adaptation network applied to unmanned aerial vehicle fault diagnosis,” Sensors, vol. 25, no. 6, Art. no. 1924, 2025, doi: 10.3390/s25061924.
Y. Li, Q. Fan, H. Huang, et al., “A modified YOLOv8 detection network for UAV aerial image recognition,” Drones, vol. 7, no. 5, Art. no. 304, 2023, doi: 10.3390/drones7050304.
Y. Zou and G. Fang, “Underwater image enhancement based on generative adversarial network with multi-attention,” in Proc. 2024 5th Int. Conf. Big Data Artif. Intell. Internet Things Eng. (ICBAIE), 2024, pp. 351–354, doi: 10.1109/ICBAIE63306.2024.11116943.
J. Zhang, Z. Liu, W. Jiang, et al., “Application of deep generative networks for SAR/ISAR: A review,” Artif. Intell. Rev., vol. 56, no. 10, pp. 11905– 11983, 2023, doi: 10.1007/s10462-023-10469-5.
X. Zhao, H. Zhang, C. Li, et al., “DVIF-Net: A small-target detection network for UAV aerial images based on visible and infrared fusion,” Remote Sens., vol. 17, no. 20, Art. no. 3411, 2025, doi: 10.3390/rs17203411.
S. N. Khonina, N. L. Kazanskiy, I. V. Oseledets, et al., “Eyes of the future: Decoding the world through machine vision,” Technologies, vol. 13, no. 11, Art. no. 507, 2025, doi: 10.3390/technologies13110507.
N. Manakitsa, G. S. Maraslidis, L. Moysis, et al., “A review of machine learning and deep learning for object detection, semantic segmentation, and human action recognition in machine and robotic vision,” Technologies, vol. 12, no. 2, Art. no. 15, 2024, doi: 10.3390/technologies12020015.