OGSA-YOLO: Omni-directional Gated Spatial-Attention Network for Conveyor Belt Monitoring in Complex Environments

Abstract Existing conveyor belt monitoring methods operating under complex conditions generally suffer from severe environmental background interference, feature conflicts between multi-scale targets, and insufficient directional boundary perception capabilities. To address these issues, this study proposes an omni-directional gated spatial attention network (OGSA-YOLO), which establishes a cohesive and unified architecture to systematically decouple multi-scale feature variations and suppress non-structural environmental noise. The framework operates as a progressive feature-refinement pipeline: first, an omni-directional perception strategy (combining ODP-Conv and GSDF-Block) collaboratively separates high-frequency directional details from macroscopic semantics. Subsequently, a structural prior-based attention mechanism (SCSAB) proactively filters out intense ambient dust while preserving fragile target boundaries. Finally, an adaptive feature fusion module (AMFAM) dynamically aligns these refined cross-scale features to resolve semantic conflicts. We conduct extensive experiments on the conveyor belt monitoring dataset and the Cityscapes dataset. On the conveyor belt dataset, the \(\textrm{mAP}_{50}\) and \(\textrm{mAP}_{95}\) for the detection task reach 96.9% and 84.8%, while the \(\textrm{mAP}_{50}\) and \(\textrm{mAP}_{95}\) for the segmentation task reach 96.1% and 77.4%, respectively. These results comprehensively outperform most state-of-the-art methods. Data availability No datasets were generated or analyzed during the current study. References Chen L, Wu L, Ren Q (2025) A multimodal data fusion-based intelligent detection method for lump coal on underground conveyor belts in smart manufacturing. J Ind Inf Int, pp 100997 Liu F, El-Mesery HS, ElMesiry AH et al (2025) Artificial intelligence predictive thermodynamic analysis and energy enhancement of conveyor belt dryer under combined heating conditions. Energy Rep 14:4797–4809 Wang M, Shen K, Tai C et al (2023) Research on fault diagnosis system for belt conveyor based on internet of things and the lightgbm model. PLoS ONE 18(3):e0277352 Yang R, Qiao T, Pang Y et al (2020) Infrared spectrum analysis method for detection and early warning of longitudinal tear of mine conveyor belt. Measurement 165:107856 Leite JR, Cavalieri DC, Prado AR (2024) Efficient monitoring of longitudinal tears in conveyor belts using 2d laser scanner and statistical methods. Measurement 227:114225 Zhang M, Jiang K, Cao Y et al (2023) A new paradigm for intelligent status detection of belt conveyors based on deep learning. Measurement 213:112735 Hou C, Qiao T, Zhang H et al (2019) Multispectral visual detection method for conveyor belt longitudinal tear. Measurement 143:246–257 Che J, Qiao T, Yang Y et al (2021) Longitudinal tear detection method of conveyor belt based on audio-visual fusion. Measurement 176:109152 Wang Y, Miao C, Liu Y et al (2022) Research on a sound-based method for belt conveyor longitudinal tear detection. Measurement 190:110787 Khanam R, Hussain M (2024) Yolov11: an overview of the key architectural enhancements. arXiv preprint arXiv:2410.17725 Emin G, Ismail HS, Serap C, Ozhan A, Erdeniz E, Mert D, Cuneyt B (2022) Electric shore-to-ship charging socket detection using image processing and yolo. In 2022 International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT), pp 1069–1073. IEEE Güney E, Bayılmış C, Çakar S, Erol E, Atmaca Ö (2024) Autonomous control of shore robotic charging systems based on computer vision. Expert Syst Appl 238:122116 Emin G, Cuneyt B (2024) Yolov5-based driver behavior monitoring system for safer roads on jetson xavier nx. In: International Conference on Advanced Engineering, Technology and Applications, pp 339–350. Springer Wu L, Zhang L, Chen L et al (2023) A lightweight and multisource information fusion method for real-time monitoring of lump coal on mining conveyor belts. Int J Intell Syst 2023(1):5327122 Kou Q, Ma H, Xu J et al (2023) Coal flow foreign body classification based on escbam and multi-channel feature fusion. Sensors 23(15):6831 Ling J, Fu Z, Yuan X (2025) Lightweight coal mine conveyor belt foreign object detection based on improved yolov8n. Sci Rep 15(1):10361 Luo B, Kou Z, Han C et al (2023) A faster and lighter detection method for foreign objects in coal mine belt conveyors. Sensors 23(14):6276 Hu F, Zhou M, Yan P et al (2022) A bayesian optimal convolutional neural network approach for classification of coal and gangue with multispectral imaging. Opt Lasers Eng 156:107081 Fan H, Liu J, Yan X et al (2024) A fast and high-accuracy foreign object detection method for belt conveyor coal flow images with target occlusion. Sensors 24(16):5251 Zhang M, Jiang K, Cao Y et al (2022) A deep learning-based method for deviation status detection in intelligent conveyor belt system. J Clean Prod, pp 132575 Xu X, Zhao H, Fu X et al (2023) Real-time belt deviation detection method based on depth edge feature and gradient constraint. Sensors 23(19):8208 Ni Y, Cheng H, Hou Y et al (2024) Study of conveyor belt deviation detection based on improved yolov8 algorithm. Sci Rep 14(1):26876 Zhao L, Guo C, Zhu Y et al (2025) A real-time detection algorithm for high-precision conveyor belt deviation applied to edge computing devices. Eng Failure Anal, pp 110480 Gao M, Li S, Chen X et al (2024) A novel combined method for conveyor belt deviation discrimination under complex operational scenarios. Eng Appl Artif Intell 137:109145 Žvirblis T, Petkevičius L, Bzinkowski D et al (2022) Investigation of deep learning models on identification of minimum signal length for precise classification of conveyor rubber belt loads. Adv Mech Eng 14(6):16878132221102776 He K, Gkioxari G, Dollár P et al (2017) Mask r-cnn. In: Proceedings of the IEEE International Conference on Computer Vision, pp 2961–2969 Cai Z, Vasconcelos N (2018) Cascade r-cnn: delving into high quality object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 6154–6162 Kirillov A, Wu Y, He K et al (2020) Pointrend: image segmentation as rendering. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 9799–9808 Chen K, Pang J, Wang J et al (2019) Hybrid task cascade for instance segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 4974–4983 Liu Z, Lin Y, Cao Y et al (2021) Swin transformer: Hierarchical vision transformer using shifted windows. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 10012–10022 Yi D, Ahmedov HB, Jiang S et al (2024) Coordinate-aware mask r-cnn with group normalization: a underwater marine animal instance segmentation framework. Neurocomputing 583:127488 Xie C, Xiang Y, Mousavian A et al (2021) Unseen object instance segmentation for robotic environments. IEEE Trans Rob 37(5):1343–1359 Zhu C, Liu K, Tang W et al (2025) Hard-aware instance adaptive self-training for unsupervised cross-domain semantic segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence Wang X, Kong T, Shen C et al (2020) Solo: segmenting objects by locations. In: European conference on computer vision, pages 649–665. Springer Wang X, Zhang R, Kong T et al (2020) Solov2: dynamic and fast instance segmentation. Adv Neural Inf Process Syst 33:17721–17732 Bolya D, Zhou C, Xiao F et al (2019) Yolact: real-time instance segmentation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp 9157–9166 Fang Y, Yang S, Wang X et al (2021) Instances as queries. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 6910–6919 Li F, Zhang H, Xu H et al (2023) Mask dino: towards a unified transformer-based framework for object detection and segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 3041–3050 Cheng B, Misra I, Schwing A G et al (2022) Masked-attention mask transformer for universal image segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 1290–1299 Li W, Liu W, Zhu J et al (2024) Box2mask: box-supervised instance segmentation via level-set evolution. IEEE Trans Pattern Anal Mach Intell 46(7):5157–5173 Tian Y, Ye Q, Doermann D (2025) Yolov12: Attention-centric real-time object detectors. arXiv preprint arXiv:2502.12524 Lei M, Li S, Wu Y et al (2025) Yolov13: Real-time object detection with hypergraph-enhanced adaptive visual perception. arXiv preprint arXiv:2506.17733 Kang M, Ting CM, Ting FF et al (2024) Asf-yolo: a novel yolo model with attentional scale sequence fusion for cell instance segmentation. Image Vis Comput 147:105057 Khan Z, Liu H, Shen Y et al (2024) Deep learning improved yolov8 algorithm: real-time precise instance segmentation of crown region orchard canopies in natural environment. Comput Electron Agric 224:109168 Ye G, Li S, Zhou M et al (2024) Pavement crack instance segmentation using yolov7-wmf with connected feature fusion. Autom Constr 160:105331 Lin TY, Dollár P, Girshick R et al (2017) Feature pyramid networks for object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 2117–2125 Doherty J, Gardiner B, Kerr E et al (2025) Bifpn-yolo: one-stage object detection integrating bi-directional feature pyramid networks. Pattern Recogn 160:111209 Zhang Y, Zhang T, Wu C et al (2023) Multi-scale spatiotemporal feature fusion network for video saliency prediction. IEEE Trans Multimedia 26:4183–4193 Wu H, Huang P, Zhang M et al (2023) Cmtfnet: Cnn and multiscale transformer fusion network for remote-sensing image semantic segmentation. IEEE Trans Geosci Remote Sens 61:1–12 Feng Y, Xu H, Jiang J et al (2022) Icif-net: intra-scale cross-interaction and inter-scale feature fusion network for bitemporal remote sensing images change detection. IEEE Trans Geosci Remote Sens 60:1–13 Liu M, Dan J, Lu Z et al (2024) Cm-unet: hybrid cnn-mamba unet for remote sensing image semantic segmentation. arXiv preprint arXiv:2405.10530 Wang Q, Huang J, Meng Y et al (2024) Df2net: differential feature fusion network for hyperspectral image classification. IEEE J Sel Topics Appl Earth Observation Remote Sens 17:10660–10673 Huo X, Sun G, Tian S et al (2024) Hifuse: Hierarchical multi-scale feature fusion network for medical image classification. Biomed Signal Process Control 87:105534 Huang M, Yu W, Zhang L (2024) Df3net: dual frequency feature fusion network with hierarchical transformer for image inpainting. Information Fusion 111:102487 He X, Li X, Li B et al (2025) Multi-modal collaborative learning with vision foundation model prompt boosts 3d semi-supervised semantic segmentation. Inf Fusion, pp 104019 Dong H, Dong J, Sun S et al (2024) Crop water stress detection based on uav remote sensing systems. Agric Water Manag 303:109059 Si Y, Huiying X, Zhu X, Zhang W, Dong Y, Chen Y, Li H (2025) Scsa: exploring the synergistic effects between spatial and channel attention. Neurocomputing 634:129866 Cordts M, Omran M, Ramos S et al (2016) The cityscapes dataset for semantic urban scene understanding. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp 3213–3223 Chengqi L, Wenwei Z, Haian H, Yue Z, Yudong W, Yanyi L, Shilong Z, Kai C (2022) Rtmdet: an empirical study of designing real-time object detectors. arXiv preprint arXiv:2212.07784 Sanghyun W, Jongchan P, Joon-Young L, Kweon In So (2018) Cbam: Convolutional block attention module. In: Proceedings of the European Conference on Computer Vision (ECCV), pp 3–19 Liu Y, Shao Z, Hoffmann N (2021) Global attention mechanism: retain information to enhance channel-spatial interactions.arXiv:2112.05561 arXiv preprint Qibin H, Daquan Z, Jiashi F (2021) Coordinate attention for efficient mobile network design. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 13713–13722 Qilong W, Banggu W, Pengfei Z, Peihua L, Wangmeng Z, Qinghua H (2020) Eca-net: efficient channel attention for deep convolutional neural networks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp 11534–11542 Author information Authors and Affiliations Contributions Kelei Sun contributed to supervision, investigation, and funding acquisition; Xuedong Feng contributed to writing—conceptualization, methodology, software, and writing—original draft; Huaping Zhou contributed to writing—review & editing; Tao Wu contributed to software; Bin Deng contributed to visualization. Corresponding author Ethics declarations Competing interests The authors declare no competing interests. Additional information Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Rights and permissions Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law. About this article Cite this article Sun, K., Feng, X., Zhou, H. et al. OGSA-YOLO: Omni-directional Gated Spatial-Attention Network for Conveyor Belt Monitoring in Complex Environments. J Supercomput 82, 706 (2026). https://doi.org/10.1007/s11227-026-08817-7 Received: Accepted: Published: Version of record: DOI: https://doi.org/10.1007/s11227-026-08817-7