BSG-Mamba: A dual-branch network based on background-suppression sparse graph-enhanced Mamba for infrared small target detection

Xinying Huang; Jianhua Song; Xinrong Fu; Yu Chen; Hao Liu; Xinying Huang; Jianhua Song; Xinrong Fu; Yu Chen; Hao Liu

doi:10.3934/era.2026272

Electronic Research Archive

2026, Volume 34, Issue 9: 6177-6218. doi: 10.3934/era.2026272

Previous Article Next Article

Research article Special Issues

BSG-Mamba: A dual-branch network based on background-suppression sparse graph-enhanced Mamba for infrared small target detection

1.
College of Physics and Information Engineering, Minnan Normal University, Zhangzhou 363000, China
2.
Key Laboratory of Light Field Manipulation and System Integration Applications in Fujian Province, Minnan Normal University, Zhangzhou 363000, China

Received: 09 May 2026 Revised: 19 June 2026 Accepted: 06 July 2026 Published: 16 July 2026

Infrared small target detection faces numerous challenges due to weak target features and complex backgrounds. Existing methods have drawbacks such as mixed background and target signals, imbalanced local and global feature modeling, and insufficient learning of direction-specific context. To address these issues, this paper proposes a dual-branch network based on background-suppression sparse graph-enhanced Mamba (BSG-Mamba). We build this network on a U-shaped encoder-decoder architecture. We design a background-suppression branch to initially separate the target from the background and combine it with a target-enhancing gating (TEG) mechanism to enhance target features. On this basis, we effectively capture non-local feature correlations in small targets by embedding a sparse graph encoder (SG-encoder), compensating for the shortcomings of traditional convolutional segmentation. At the deep encoder position, it constructs a Mamba module with multi-state space model (SSM) decoupling and adaptive fusion. We configure independent SSM parameters for four-direction scanning and introduce learnable fusion weights to achieve unified modeling of long-range dependencies and direction-specific features. Experimental results on the NUAA-SIRST and NUDT-SIRST datasets show that the core detection metrics of the model outperform existing state-of-the-art (SOTA) methods, achieving detection rates of 97.74% and 98.90%, respectively, on the two datasets, while maintaining a lightweight network parameter count and balancing detection accuracy and computational efficiency.
- infrared small target detection,
- background suppression,
- Unet,
- sparse graph,
- Mamba
Citation: Xinying Huang, Jianhua Song, Xinrong Fu, Yu Chen, Hao Liu. BSG-Mamba: A dual-branch network based on background-suppression sparse graph-enhanced Mamba for infrared small target detection[J]. Electronic Research Archive, 2026, 34(9): 6177-6218. doi: 10.3934/era.2026272

Related Papers:

Abstract

Infrared small target detection faces numerous challenges due to weak target features and complex backgrounds. Existing methods have drawbacks such as mixed background and target signals, imbalanced local and global feature modeling, and insufficient learning of direction-specific context. To address these issues, this paper proposes a dual-branch network based on background-suppression sparse graph-enhanced Mamba (BSG-Mamba). We build this network on a U-shaped encoder-decoder architecture. We design a background-suppression branch to initially separate the target from the background and combine it with a target-enhancing gating (TEG) mechanism to enhance target features. On this basis, we effectively capture non-local feature correlations in small targets by embedding a sparse graph encoder (SG-encoder), compensating for the shortcomings of traditional convolutional segmentation. At the deep encoder position, it constructs a Mamba module with multi-state space model (SSM) decoupling and adaptive fusion. We configure independent SSM parameters for four-direction scanning and introduce learnable fusion weights to achieve unified modeling of long-range dependencies and direction-specific features. Experimental results on the NUAA-SIRST and NUDT-SIRST datasets show that the core detection metrics of the model outperform existing state-of-the-art (SOTA) methods, achieving detection rates of 97.74% and 98.90%, respectively, on the two datasets, while maintaining a lightweight network parameter count and balancing detection accuracy and computational efficiency.

References

[1]	M. Teutsch, W. Krüger, Classification of small boats in infrared images for maritime surveillance, in 2010 International WaterSide Security Conference, (2010), 1-7. https://doi.org/10.1109/WSSC.2010.5730289
[2]	C. Li, Z. Huang, X. Xie, W. Li, IST-TransNet: Infrared small target detection based on transformer network, Infrared Phys. Technol. , 132 (2023), 104723. https://doi.org/10.1016/j.infrared.2023.104723 doi: 10.1016/j.infrared.2023.104723
[3]	H. Yi, C. Yang, R. Qie, J. Liao, F. Wu, T. Pu, et al., Spatial-temporal tensor ring norm regularization for infrared small target detection, IEEE Geosci. Remote Sens. Lett. , 20 (2023), 1-5. https://doi.org/10.1109/LGRS.2023.3236030 doi: 10.1109/LGRS.2023.3236030
[4]	J. Yao, S. Xu, H. Feijiang, C. Su, Improved lightweight infrared road target detection method based on YOLOv8, Infrared Phys. Technol. , 141 (2024), 105497. https://doi.org/10.1016/j.infrared.2024.105497 doi: 10.1016/j.infrared.2024.105497
[5]	S. Li, B. Hao, Adversarial camouflage for mobile targets against battlefield optical and infrared AI detection, J. King Saud Univ. Comput. Inf. Sci., 38 (2026), 135. https://doi.org/10.1007/s44443-026-00542-8
[6]	N. Kumar, P. Singh, Small and dim target detection in infrared imagery: A review, current techniques and future directions, Neurocomputing, 630 (2025), 129640. https://doi.org/10.1016/j.neucom.2025.129640
[7]	X. Bai, F. Zhou, Analysis of new top-hat transformation and the application for infrared dim small target detection, Pattern Recognit. , 43 (2010), 2145-2156. https://doi.org/10.1016/j.patcog.2009.12.023 doi: 10.1016/j.patcog.2009.12.023
[8]	S. D. Deshpande, M. H. Er, R. Venkateswarlu, P. Chan, Max-mean and max-median filters for detection of small targets, in Signal and Data Processing of Small Targets 1999, 3809 (1999), 74-83. https://doi.org/10.1117/12.364049
[9]	Y. Li, Z. Li, C. Zhang, Z. Luo, Y. Zhu, Z. Ding, et al., Infrared maritime dim small target detection based on spatiotemporal cues and directional morphological filtering, Infrared Phys. Technol. , 115 (2021), 103657. https://doi.org/10.1016/j.infrared.2021.103657 doi: 10.1016/j.infrared.2021.103657
[10]	C. L. P. Chen, H. Li, Y. Wei, T. Xia, Y. Y. Tang, A local contrast method for small infrared target detection, IEEE Trans. Geosci. Remote Sens. , 52 (2014), 574-581. https://doi.org/10.1109/TGRS.2013.2242477 doi: 10.1109/TGRS.2013.2242477
[11]	Y. Wei, X. You, H. Li, Multiscale patch-based contrast measure for small infrared target detection, Pattern Recognit. , 58 (2016), 216-226. https://doi.org/10.1016/j.patcog.2016.04.002 doi: 10.1016/j.patcog.2016.04.002
[12]	X. Wang, G. Lv, L. Xu, Infrared dim target detection based on visual attention, Infrared Phys. Technol. , 55 (2012), 513-521. https://doi.org/10.1016/j.infrared.2012.08.004 doi: 10.1016/j.infrared.2012.08.004
[13]	H. Zhu, H. Ni, S. Liu, G. Xu, L. Deng, TNLRS: Target-aware non-local low-rank modeling with saliency filtering regularization for infrared small target detection, IEEE Trans. Image Process. , 29 (2020), 9546-9558. https://doi.org/10.1109/TIP.2020.3028457 doi: 10.1109/TIP.2020.3028457
[14]	O. Ronneberger, P. Fischer, T. Brox, U-Net: Convolutional networks for biomedical image segmentation, in Medical Image Computing and Computer Assisted Intervention (MICCAI), (2015), 234-241. https://doi.org/10.1007/978-3-319-24574-4_28
[15]	M. Liu, H. Du, Y. Zhao, L. Dong, M. Hui, Image small target detection based on deep learning with SNR controlled sample generation, Curr. Trends Comput. Sci. Mech. Autom. , 1 (2022), 211-220. https://doi.org/10.1515/ 9783110584974-025 doi: 10.1515/9783110584974-025
[16]	Y. Dai, Y. Wu, F. Zhou, K. Barnard, Asymmetric contextual modulation for infrared small target detection, preprint, arXiv: 2009.14530. https://doi.org/10.48550/arXiv.2009.14530
[17]	T. Zhang, L. Li, S. Cao, T. Pu, Z. Peng, Attention-guided pyramid context networks for detecting infrared small target under complex background, IEEE Trans. Aerosp. Electron. Syst. , 59 (2023), 4250-4261. https://doi.org/10.1109/TAES.2023.3238703 doi: 10.1109/TAES.2023.3238703
[18]	X. Wu, D. Hong, J. Chanussot, UIU-Net: U-Net in U-Net for infrared small object detection, IEEE Trans. Image Process. , 32 (2023), 364-376. https://doi.org/10.1109/TIP.2022.3228497 doi: 10.1109/TIP.2022.3228497
[19]	B. Li, C. Xiao, L. Wang, Y. Wang, Z. Lin, M. Li, et al., Dense nested attention network for infrared small target detection, preprint, arXiv: 2106.00487. https://doi.org/10.48550/arXiv.2106.00487
[20]	J. Lin, K. Zhang, X. Yang, X. Cheng, C. Li, Infrared dim and small target detection based on U-Transformer, J. Vis. Commun. Image Represent. , 89 (2022), 103684. https://doi.org/10.1016/j.jvcir.2022.103684 doi: 10.1016/j.jvcir.2022.103684
[21]	T. Wu, B. Li, Y. Luo, Y. Wang, C. Xiao, T. Liu, et al., MTU-Net: Multilevel TransUNet for space-based infrared tiny ship detection, IEEE Trans. Geosci. Remote Sens. , 61 (2023), 1-15. https://doi.org/10.1109/TGRS.2023.3235002 doi: 10.1109/TGRS.2023.3235002
[22]	S. Yuan, H. Qin, X. Yan, N. Akhtar, A. Mian, SCTransNet: Spatial-channel cross transformer network for infrared small target detection, IEEE Trans. Geosci. Remote Sens. , 62 (2024), 1-15. https://doi.org/10.1109/TGRS.2024.3383649 doi: 10.1109/TGRS.2024.3383649
[23]	Y. Zhang, W. Bao, W. Wan, Q. Xiao, Y. Tang, X. Zou, et al., APTNet: Adaptive partial transformer network for infrared small target detection, IEEE Sensors J. , 25 (2025), 17960-17974. https://doi.org/10.1109/JSEN.2025.3559093 doi: 10.1109/JSEN.2025.3559093
[24]	H. Wu, X. Huang, C. He, H. Xiao, S. Luo, Infrared small target detection with Swin transformer-based multiscale atrous spatial pyramid pooling network, IEEE Trans. Instrum. Meas. , 74 (2025), 1-14. https://doi.org/10.1109/TIM.2024.3522414 doi: 10.1109/TIM.2024.3522414
[25]	M. Zhang, R. Zhang, Y. Yang, H. Bai, J. Zhang, J. Guo, ISNet: Shape matters for infrared small target detection, in 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), (2022), 867-876. https://doi.org/10.1109/CVPR52688.2022.00095
[26]	T. Liu, Y. Liu, J. Yang, B. Li, Y. Wang, W. An, Graph Laplacian regularization for fast infrared small target detection, Pattern Recognit., 158 (2025), 111077. https://doi.org/10.1016/j.patcog.2024.111077
[27]	A. Gu, K. Goel, C. Ré, Efficiently modeling long sequences with structured state spaces, preprint, arXiv: 2111.00396. https://doi.org/10.48550/arXiv.2111.00396
[28]	A. Gu, T. Dao, Mamba: Linear-time sequence modeling with selective state spaces, preprint, arXiv: 2312.00752. https://doi.org/10.48550/arXiv.2312.00752
[29]	J. Ma, F. Li, B. Wang, U-Mamba: Enhancing long-range dependency for biomedical image segmentation, preprint, arXiv: 2401.04722. https://doi.org/10.48550/arXiv.2401.04722
[30]	J. Xu, HC-Mamba: Vision MAMBA with hybrid convolutional techniques for medical image segmentation, preprint, arXiv: 2405.05007. https://doi.org/10.48550/arXiv.2405.05007
[31]	R. Zhao, S. H. Tang, J. Shen, E. E. B. Supeni, S. A. Rahim, Enhancing autonomous driving safety: A robust traffic sign detection and recognition model TSD-YOLO, Signal Process. , 225 (2024), 109619. https://doi.org/10.1016/j.sigpro.2024.109619 doi: 10.1016/j.sigpro.2024.109619
[32]	J. Han, K. Liang, B. Zhou, X. Zhu, J. Zhao, L. Zhao, Infrared small target detection utilizing the multiscale relative local contrast measure, IEEE Geosci. Remote Sens. Lett. , 15 (2018), 612-616. https://doi.org/10.1109/LGRS.2018.2790909 doi: 10.1109/LGRS.2018.2790909
[33]	K. Wang, S. Du, C. Liu, Z. Cao, Interior attention-aware network for infrared small target detection, IEEE Trans. Geosci. Remote Sens. , 60 (2022), 1-13. https://doi.org/10.1109/TGRS.2022.3163410 doi: 10.1109/TGRS.2022.3163410
[34]	X. Ling, L. Wu, C. Wu, S. Ji, Graph neural networks: Graph matching, in Graph Neural Networks: Foundations, Frontiers, and Applications, (2022), 277-295. https://doi.org/10.1007/978-981-16-6054-2_13
[35]	B. Jiang, Z. Zhang, D. Lin, J. Tang, B. Luo, Semi-supervised learning with graph learning-convolutional networks, in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), (2019), 11305-11312. https://doi.org/10.1109/CVPR.2019.01157
[36]	P. Veličković, A. Casanova, P. Liò, G. Cucurull, A. Romero, Y. Bengio, Graph attention networks, in 6th International Conference on Learning Representations (ICLR), 2018. https://doi.org/10.17863/CAM.48429
[37]	Y. Chen, M. Rohrbach, Z. Yan, S. Yan, J. Feng, Y. Kalantidis, Graph-based global reasoning networks, in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), (2019), 433-442. https://doi.org/10.1109/CVPR.2019.00052
[38]	K. Han, Y. Wang, J. Guo, Y. Tang, E. Wu, Vision GNN: An image is worth graph of nodes, preprint, arXiv: 2206.00272. https://doi.org/10.48550/arXiv.2206.00272
[39]	M. Munir, W. Avery, R. Marculescu, MobileViG: Graph-based sparse attention for mobile vision applications, in 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), (2023), 2211-2219. https://doi.org/10.1109/CVPRW59228.2023.00215
[40]	J. Lin, Y. Ruan, S. Li, X. Yang, S. Niu, Z. Meng, BPGA: Enhancing infrared small and dim target detection with background prior guide and vision graph attention, Opt. Laser Technol. , 192 (2025), 113867. https://doi.org/10.1016/j.optlastec.2025.113867 doi: 10.1016/j.optlastec.2025.113867
[41]	Y. Shen, Q. Li, C. Xu, C. Chang, Q. Yin, Graph-based context learning network for infrared small target detection, Neurocomputing, 616 (2025), 128949. https://doi.org/10.1016/j.neucom.2024.128949 doi: 10.1016/j.neucom.2024.128949
[42]	T. Chen, Z. Ye, Z. Tan, T. Gong, Y. Wu, Q. Chu, et al., MiM-ISTD: Mamba-in-Mamba for efficient infrared small-target detection, IEEE Trans. Geosci. Remote Sens. , 62 (2024), 1-13. https://doi.org/10.1109/TGRS.2024.3485721 doi: 10.1109/TGRS.2024.3485721
[43]	R. E. Kalman, A new approach to linear filtering and prediction problems, J. Basic Eng. , 82 (1960), 35-45. https://doi.org/10.1115/1.3662552 doi: 10.1115/1.3662552
[44]	L. Zhu, B. Liao, Q. Zhang, X. Wang, W. Liu, X. Wang, Vision Mamba: Efficient visual representation learning with bidirectional state space model, preprint, arXiv: 2401.09417. https://doi.org/10.48550/arXiv.2401.09417
[45]	H. Guo, J. Li, T. Dai, Z. Ouyang, X. Ren, S. T. Xia, MambaIR: A simple baseline for image restoration with state-space model, preprint, arXiv: 2402.15648. https://doi.org/10.48550/arXiv.2402.15648
[46]	K. Chen, B. Chen, C. Liu, W. Li, Z. Zou, Z. Shi, RSMamba: Remote sensing image classification with state space model, IEEE Geosci. Remote Sens. Lett. , 21 (2024), 1-5. https://doi.org/10.1109/LGRS.2024.3407111 doi: 10.1109/LGRS.2024.3407111
[47]	W. Liu, X. Lu, J. Zhang, D. Li, X. Zhang, MOU-Mamba: Multi-order U-shape Mamba for infrared small target detection, Opt. Laser Technol. , 187 (2025), 112851. https://doi.org/10.1016/j.optlastec.2025.112851 doi: 10.1016/j.optlastec.2025.112851
[48]	D. Hong, B. Zhang, X. Li, Y. Li, C. Li, J. Yao, et al., SpectralGPT: Spectral remote sensing foundation model, IEEE Trans. Pattern Anal. Mach. Intell. , 46 (2024), 5227-5244. https://doi.org/10.1109/TPAMI.2024.3362475 doi: 10.1109/TPAMI.2024.3362475
[49]	C. Li, B. Zhang, D. Hong, J. Yao, J. Chanussot, LRR-Net: An interpretable deep unfolding network for hyperspectral anomaly detection, IEEE Trans. Geosci. Remote Sens. , 61 (2023), 1-12. https://doi.org/10.1109/TGRS.2023.3279834 doi: 10.1109/TGRS.2023.3279834
[50]	C. Gao, D. Meng, Y. Yang, Y. Wang, X. Zhou, A. G. Hauptmann, Infrared patch-image model for small target detection in a single image, IEEE Trans. Image Process. , 22 (2013), 4996-5009. https://doi.org/10.1109/TIP.2013.2281420 doi: 10.1109/TIP.2013.2281420
[51]	T. Zhang, H. Wu, Y. Liu, L. Peng, C. Yang, Z. Peng, Infrared small target detection based on non-convex optimization with Lp-norm constraint, Remote Sens. , 11 (2019), 559. https://doi.org/10.3390/rs11050559 doi: 10.3390/rs11050559

Reader Comments

Your name:*

Email:*
© 2026 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)