AI Chat Paper
Note: Please note that the following content is generated by AMiner AI. SciOpen does not take any responsibility related to this content.
{{lang === 'zh_CN' ? '文章概述' : 'Summary'}}
{{lang === 'en_US' ? '中' : 'Eng'}}
Chat more with AI
Powered by AMiner. Clicking this button will take you away from SciOpen.
Home Chinese Journal of Aeronautics Article
Article Link
Cite
EndNote(RIS) BibTeX
Collect
Show Outline
Outline
Show full outline
Hide outline
Outline
Show full outline
Hide outline
Full Length Article | Open Access

Few-shot incremental radar target recognition framework based on scattering-topology properties

Chenxuan LIa, Weigang ZHUb,( ), Bakun ZHUa, Yonggang LIa
Graduate School, Space Engineering University, Beijing 101400, China
Department of Electronic and Optical Engineering, Space Engineering University, Beijing 101400, China

Peer review under responsibility of Editorial Committee of CJA.

Show Author Information

Abstract

The continuous emergence of new targets in open scenarios leads to a substantial decrease in the performance of Inverse Synthetic Aperture Radar (ISAR) recognition systems. Also, data scarcity further exacerbates the challenge of identifying new classes of ISAR targets. In this paper, a few-shot incremental target recognition framework based on Scattering-Topology Properties (STPIL) is proposed. Specifically, STPIL extracts scattering-topology properties of ISAR targets as recognition features. Meanwhile, the pseudo-incremental training strategy effectively alleviates the algorithm’s forgetting of old knowledge, and improves compatibility with new classes. Besides, a feature embedding network, with few parameters, is designed based on the graph neural network. This embedding network is highly adaptable to changes in data distribution. Additionally, STPIL fully considers the joint distribution and marginal distribution in scattering features, and uses the Brownian distance metric module to make the scattering-topology features more discriminative. Experimental results on both the simulation dataset and the public measured data indicate that STPIL can effectively balance new classes with old classes, and has superior performance to other advanced methods in the incremental recognition of targets.

Keywords

ScatteringGraph neural networksIncremental learningInverse Synthetic Aperture Radar (ISAR)Brownian distance metric

References

1

Zhou XN, Bai XR, Wang L, et al. Robust ISAR target recognition based on ADRISAR-net. IEEE Trans Aerosp Electron Syst 2022;58(6):5494–505.
Crossref Google Scholar
2

Lu W, Zhang YS, Yin CB, et al. A deformation robust ISAR image satellite target recognition method based on PT-CCNN. IEEE Access 2021;9:23432–53.
Crossref Google Scholar
3

Zhang Y, Yuan HX, Li HB, et al. Meta-learner-based stacking network on space target recognition for ISAR images. IEEE J Sel Top Appl Earth Obs Remote Sens 2021;14:12132–48.
Crossref Google Scholar
4

Liu D, Li JX. Spectral-spatial target detection based on data field modeling for hyperspectral data. Chin J Aeronaut 2018;31(4):795–805.
Crossref Google Scholar
5

Li RZ, Zhang SH, Zhang C, et al. Deep learning approach for sparse aperture ISAR imaging and autofocusing based on complex-valued ADMM-Net. IEEE Sens J 2021;21(3):3437–51.
Crossref Google Scholar
6
Rebuffi SA, Kolesnikov A, Sperl G, et al. iCaRL: Incremental classifier and representation learning. 2017 IEEE conference on computer vision and pattern recognition (CVPR). Piscataway: IEEE Press; 2017. p. 5533–42.
Crossref
7
Hersche M, Karunaratne G, Cherubini G, et al. Constrained few-shot class-incremental learning. 2022 IEEE/CVF conference on computer vision and pattern recognition (CVPR). Piscataway: IEEE Press; 2022. p. 9047–57.
Crossref
8
Zhou DW, Wang FY, Ye HJ, et al. Forward compatible fewshot class-incremental learning. 2022 IEEE/CVF conference on computer vision and pattern recognition (CVPR). Piscataway: IEEE Press; 2022. p. 9036–46.
Crossref
9
Wang ZF, Zhang ZZ, Ebrahimi S, et al. DualPrompt: Complementary prompting for rehearsal-free continual learning. European conference on computer vision. Cham: Springer; 2022. p. 631–48.
Crossref
10
Zhang C, Song N, Lin GS, et al. Few-shot incremental learning with continually evolved classifiers. 2021 IEEE/CVF conference on computer vision and pattern recognition (CVPR). Piscataway: IEEE Press; 2021. p. 12450–9.
Crossref
11

Masana M, Liu XL, Twardowski B, et al. Class-incremental learning: Survey and performance evaluation on image classification. IEEE Trans Pattern Anal Mach Intell 2023;45(5):5513–33.
Crossref Google Scholar
12

Shan LL, Wang WQ, Lv K, et al. Class-incremental learning for semantic segmentation in aerial imagery via distillation in all aspects. IEEE Trans Geosci Remote Sens 2021;60:5615712.
Crossref Google Scholar
13

Dang SH, Cui ZY, Cao ZJ, et al. Distribution reliability assessment-based incremental learning for automatic target recognition. IEEE Trans Geosci Remote Sens 2023;61:5208413.
Crossref Google Scholar
14

Lu XN, Sun X, Diao WH, et al. LIL: Lightweight incremental learning approach through feature transfer for remote sensing image scene classification. IEEE Trans Geosci Remote Sens 2021;60:5611320.
Crossref Google Scholar
15

Deng JR, Zhang HJ, Hu JH, et al. Class incremental robotic pick-and-place via incremental few-shot object detection. IEEE Robot Autom Lett 2023;8(9):5974–81.
Crossref Google Scholar
16

Li DW, Huang H. Few-shot class-incremental learning via compact and separable features for fine-grained vehicle recognition. IEEE Trans Intell Transp Syst 2022;23(11):21418–29.
Crossref Google Scholar
17

Bai J, Yuan AR, Xiao Z, et al. Class incremental learning with few-shots based on linear programming for hyperspectral image classification. IEEE Trans Cybern 2022;52(6):5474–85.
Crossref Google Scholar
18

Pal D, Bose S, Banerjee B, et al. Extreme value meta-learning for few-shot open-set recognition of hyperspectral images. IEEE Trans Geosci Remote Sens 2023;61:5512516.
Crossref Google Scholar
19

Liu BH, Yang BY, Xie LX, et al. Learnable distribution calibration for few-shot class-incremental learning. IEEE Trans Pattern Anal Mach Intell 2023;45(10):12699–706.
Crossref Google Scholar
20

Cheng M, Wang HL, Long Y. Meta-learning-based incremental few-shot object detection. IEEE Trans Circuits Syst Video Technol 2022;32(4):2158–69.
Crossref Google Scholar
21

Tai Y, Tan YH, Xiong SZ, et al. Mine-distill-prototypes for complete few-shot class-incremental learning in image classification. IEEE Trans Geosci Remote Sens 2023;61:5206013.
Crossref Google Scholar
22

Yang BY, Lin MB, Zhang YX, et al. Dynamic support network for few-shot class incremental learning. IEEE Trans Pattern Anal Mach Intell 2023;45(3):2945–51.
Crossref Google Scholar
23

Wang L, Yang XY, Tan HY, et al. Few-shot class-incremental SAR target recognition based on hierarchical embedding and incremental evolutionary network. IEEE Trans Geosci Remote Sens 2023;61:5204111.
Crossref Google Scholar
24

Tang JX, Xiang DL, Zhang F, et al. Incremental SAR automatic target recognition with error correction and high plasticity. IEEE J Sel Top Appl Earth Obs Remote Sens 2022;15:1327–39.
Crossref Google Scholar
25

Li B, Cui ZY, Cao ZJ, et al. Incremental learning based on anchored class centers for SAR automatic target recognition. IEEE Trans Geosci Remote Sens 2022;60:5235313.
Crossref Google Scholar
26

Yu XL, Dong FL, Ren HH, et al. Multilevel adaptive knowledge distillation network for incremental SAR target recognition. IEEE Geosci Remote Sens Lett 2023;20:4004405.
Crossref Google Scholar
27

Tao L, Jiang X, Li Z, et al. Multiscale incremental dictionary learning with label constraint for SAR object recognition. IEEE Geosci Remote Sens Lett 2019;16(1):80–4.
Crossref Google Scholar
28

Zhao Y, Zhao LJ, Ding D, et al. Few-shot class-incremental SAR target recognition via cosine prototype learning. IEEE Trans Geosci Remote Sens 2023;61:5212718.
Crossref Google Scholar
29

Liu ZG, Hui XM, Fu YM. A new progressive open-set recognition method with adaptive probability threshold. Chin J Aeronaut 2023;36(11):297–310.
Crossref Google Scholar
30

Chen BJ, Shen CQ, Shi JJ, et al. Continual learning fault diagnosis: A dual-branch adaptive aggregation residual network for fault diagnosis with machine increments. Chin J Aeronaut 2023;36(6):361–77.
Crossref Google Scholar
31

Székely GJ, Rizzo ML. Brownian distance covariance. Ann Appl Stat 2009;3(4):1236–65.
Crossref Google Scholar
32
NASA 3D resource [Internet]. [cited 06 Aug 2023]. Available from: https://nasa3d.arc.nasa.gov/models.
33
Tao XY, Hong XP, Chang XY, et al. Few-shot class-incremental learning. 2020 IEEE/CVF conference on computer vision and pattern recognition (CVPR). Piscataway: IEEE Press; 2020. p. 12180–9.
Crossref
34
Xu K, Hu WH, Leskovec J, et al. How powerful are graph neural networks? arXiv prerint:1810.00826; 2018.
35
Mikhail B, Daniel H, Siyuan M, et al. Reconciling modern machine learning and the bias-variance trade-off. arXiv preprint: 1812.11118; 2018.
36
Nakkiran P, Kaplun G, Bansal Y, et al. Deep double descent: Where bigger models and more data hurt. arXiv preprint:1912.02292; 2019.
37

Luo Y, Yin LC, Bai WC, et al. An appraisal of incremental learning methods. Entropy 2020;22(11):1190.
Crossref Google Scholar
38

Zhang Y, Yuan HX, Li HB, et al. Complex-valued graph neural network on space target classification for defocused ISAR images. IEEE Geosci Remote Sens Lett 2022;19:4512905.
Crossref Google Scholar
39
Sokar G, Mocanu DC, Pechenizkiy M. Self-attention meta-learner for continual learning. AAMAS 21: Proceedings of the 20th international conference on autonomous agents and multiagent systems adaptive agents and multi-agent systems. New York: ACM; 2021. p. 1658–60.
Chinese Journal of Aeronautics
Volume 37 Issue 8,
August 2024
Pages 246-260
DOI: 10.1016/j.cja.2024.05.047
Cite this article:
LI C, ZHU W, ZHU B, et al. Few-shot incremental radar target recognition framework based on scattering-topology properties. Chinese Journal of Aeronautics, 2024, 37(8): 246-260. https://doi.org/10.1016/j.cja.2024.05.047

27

Views

0

Crossref

0

Web of Science

0

Scopus

Altmetrics
Received: 22 August 2023
Revised: 07 October 2023
Accepted: 18 April 2024
Published: 04 June 2024
© 2024 Chinese Society of Aeronautics and Astronautics.

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Return