Diffusion-augmented YOLO26-Swin cascaded framework with hybrid SHAP-CAM for autonomous power grid inspection

Abstract

Deep learning-based autonomous inspection of power grid insulators is challenged by data imbalance and model opacity. This paper presents an end-to-end solution integrating advanced data synthesis, detection, classification, and explainability. First, a conditional diffusion model generates realistic synthetic fault images to balance the dataset. A two-stage architecture based on You Only Look Once version 26 (YOLO26) extra-large and Shifted windows (Swin)-V2-B, called YOLO26-Swin, fine-tuned with Bayesian optimization, performs robust insulator detection and then fault classification. Finally, a novel SHapley Additive exPlanations with Class Activation Mapping (SHAP-CAM) method provides intuitive visual explanations for model predictions. Extensive experiments validate our framework’s superiority: it achieves an F1-score of 0.98149 and a mean Average Precision (mAP)@[0.5] of 0.98951, exceeding leading detection and classification models. This work highlights the efficacy of diffusion models for data augmentation in critical infrastructure and advances the interpretability of vision-based inspection systems.

Keywords

Bayesian optimization / Diffusion models / Generative artificial intelligence / Explainable artificial intelligence

References

[1] Seman L.O., Buratto W.G., Gonzalez G.V., Leithardt V.R.Q., Nied A., Stefenon S.F.. Differentiable neural search architecture with zero-cost metrics for insulator fault prediction. Results Eng., 2026, 29. ArticleID: 109716

[2] Stefenon S.F., Seman L.O., Sopelsa Neto N.F., Meyer L.H., Mariani V.C., Coelho L.d.S.. Group method of data handling using Christiano-Fitzgerald random walk filter for insulator fault prediction. Sensors, 2023, 23(13): 6118.

[3] Xuan Y., Zhang X.Y., Li C., Wang H., Chaoxu M.. LAM-YOLOv11 for UAV transmission line inspection: overcoming environmental challenges with enhanced detection efficiency. Multimed. Syst., 2025, 31(110): 1-13.

[4] Zhang Q., Zhang J., Li Y., Zhu C., Wang G.. ID-YOLO: a multimodule optimized algorithm for insulator defect detection in power transmission lines. IEEE Trans. Instrum. Meas., 2025, 74: 1-11.

[5] Manninen H., Ramlal C.J., Singh A., Rocke S., Kilter J., Landsberg M.. Toward automatic condition assessment of high-voltage transmission infrastructure using deep learning techniques. Int. J. Electr. Power Energy Syst., 2021, 128. ArticleID: 106726

[6] Mishra M., Singh J.G.. A comprehensive review on deep learning techniques in power system protection: trends, challenges, applications and future directions. Results Eng., 2025, 25. ArticleID: 103884

[7] Aitelhaj R., Benelmostafa B.-E., Medromi H.. Exploring the generalizability of recent object detection models in identifying defective glass insulators for UAV power line inspection a case study in Morocco. International Conference on Advanced Intelligent Systems for Sustainable Development, 2023. Marrakech, Springer: 301-311.

[8] Falahatnejad S., Karami A., Nezamabadi-pour H.. PTSRGAN: power transmission lines single image super-resolution using a generative adversarial network. Int. J. Electr. Power Energy Syst., 2024, 155. ArticleID: 109607

[9] Stefenon S.F., Cristoforetti M., Cimatti A.. Conditional diffusion to enhance performance of object detection in unbalanced data engineering drawings. Neural Comput. Appl., 2025, 37: 20335-20364.

[10] Dhariwal P., Nichol A.. Diffusion models beat GANs on image synthesis. Adv. Neural Inf. Process. Syst., 2021, 34: 8780-8794

[11] Y. Tian, Q. Ye, D. Doermann, YOLOv12: attention-centric real-time object detectors, 1, 1–13 (2025). https://doi.org/10.48550/arXiv.2502.12524. Preprint. arXiv:2502.12524

[12] Huang Z., Wang H., Tang Y., Gao F., Du B., Wang J.. A two-stage location-allocation optimization method for fixed UAV nests in power inspection considering node failure scenarios. Sensors, 2025, 25(4): 1089.

[13] Yeh H.-G., Nguyen R., Kwon S.-C.. Object classification with YOLOv5 for electric utility asset inspection using UAVs. International Conference on Artificial Intelligence in Information and Communication, 2024. Osaka, IEEE: 820-825.

[14] Cha Y.-J., Ali R., Lewis J., Büyükoztürk O.. Deep learning-based structural health monitoring. Autom. Constr., 2024, 161. ArticleID: 105328

[15] Zhou M., Wu T., Xia Z., He B., Bing Kong L., Su H.. Research progress in deep learning for ceramics surface defect detection. Measurement, 2025, 242. ArticleID: 115956

[16] Jiang P., Ergu D., Liu F., Cai Y., Ma B.. A review of YOLO algorithm developments. Proc. Comput. Sci., 2022, 199: 1066-1073.

[17] J. Redmon, S. Divvala, R. Girshick, A. Farhadi, You only look once: unified, real-time object detection (2015). https://doi.org/10.48550/arXiv.1506.02640. arXiv preprint. arXiv:1506.02640

[18] Redmon J., Divvala S., Girshick R., Farhadi A.. You only look once: unified, real-time object detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016. Las Vegas, IEEE: 779-788.

[19] Y. Tian, Q. Ye, D. Doermann, YOLOv12: attention-centric real-time object detectors (2025). https://doi.org/10.48550/arXiv.2502.12524. arXiv preprint. arXiv:2502.12524

[20] Liu J., Hu M., Dong J., Lu X.. Summary of insulator defect detection based on deep learning. Electr. Power Syst. Res., 2023, 224. ArticleID: 109688

[21] R. Sapkota, R.H. Cheppally, A. Sharda, M. Karkee, YOLO26: key architectural enhancements and performance benchmarking for real-time object detection, arXiv preprint (2026). https://doi.org/10.48550/arXiv.2509.25164. arXiv:2509.25164

[22] Carvalho J.P.M., Stefenon S.F., Leithardt V.R.Q., Seman L.O., Yow K.-C., Santana J.F.D.P.. Input attention, squeeze and excitation, and spatial transformer of YOLO for fault detection using UAV. Ain Shams Eng. J., 2026, 17(3. ArticleID: 104067

[23] N. Jegham, C.Y. Koh, M. Abdelatti, A. Hendawi, YOLO Evolution: a comprehensive benchmark and architectural review of YOLOv12, YOLO11, and their previous versions, arXiv preprint (2024). https://doi.org/10.48550/arXiv.2411.0020. arXiv:2411.0020

[24] Stefenon S.F., Singh G., Souza B.J., Freire R.Z., Yow K.-C.. Optimized hybrid YOLOu-quasi-ProtoPNet for insulators classification. IET Gener. Transm. Distrib., 2023, 17(15): 3501-3511.

[25] Song J., Qin X., Lei J., Zhang J., Wang Y., Zeng Y.. A fault detection method for transmission line components based on synthetic dataset and improved YOLOv5. Int. J. Electr. Power Energy Syst., 2024, 157. ArticleID: 109852

[26] Sadykova D., Pernebayeva D., Bagheri M., James A.. IN-YOLO: real-time detection of outdoor high voltage insulators using UAV imaging. IEEE Trans. Power Deliv., 2020, 35(3): 1599-1601.

[27] Cao Z., Chen K., Chen J., Chen Z., Zhang M.. CACS-YOLO: a lightweight model for insulator defect detection based on improved YOLOv8m. IEEE Trans. Instrum. Meas., 2024, 73: 1-10.

[28] Ding L., Rao Z.Q., Ding B., Li S.J.. Research on defect detection method of railway transmission line insulators based on GC-YOLO. IEEE Access, 2023, 11: 102635-102642.

[29] Yi W., Ma S., Li R.. Insulator and defect detection model based on improved YOLO-S. IEEE Access, 2023, 11: 93215-93226.

[30] Liu Y., Huang X., Liu D.. Weather-domain transfer-based attention YOLO for multi-domain insulator defect detection and classification in UAV images. Entropy, 2024, 26(2): 136.

[31] Stefenon S.F., Seman L.O., Klaar A.C.R., Ovejero R.G., Leithardt V.R.Q.. Hypertuned-YOLO for interpretable distribution power grid fault location based on EigenCAM. Ain Shams Eng. J., 2024, 15(6. ArticleID: 102722

[32] Jiao R., Liu J., Li K., Qiao R., Liu Y., Zhang W.. YOLO-DTAD: dynamic task alignment detection model for multi-category power defects image. IEEE Trans. Instrum. Meas., 2025, 74: 1-14.

[33] Wang Y., Song X., Feng L., Zhai Y., Zhao Z., Zhang S., Wang Q.. MCI-GLA plug-in suitable for YOLO series models for transmission line insulator defect detection. IEEE Trans. Instrum. Meas., 2024, 73: 1-12.

[34] Li J., Zhou H., Lv G., Chen J.. A2MADA-YOLO: attention alignment multiscale adversarial domain adaptation YOLO for insulator defect detection in generalized foggy scenario. IEEE Trans. Instrum. Meas., 2025, 74: 1-19.

[35] He M., Qin L., Deng X., Liu K.. MFI-YOLO: multi-fault insulator detection based on an improved YOLOv8. IEEE Trans. Power Deliv., 2024, 39(1): 168-179.

[36] Jing Y., Huang T., Gao L., Deng J.. Insulator detection based on FA-YOLO network with improved feature extraction ability. IET Image Process., 2024, 18(12): 3600-3616.

[37] Li Z., Qin Q., Yang Y., Mai X., Ieiri Y., Yoshie O.. An enhanced substation equipment detection method based on distributed federated learning. Int. J. Electr. Power Energy Syst., 2025, 166. ArticleID: 110547

[38] Li S., Wang M., Zhou Y., Su Q., Liu L., Wu T.. IG-YOLOv8: insulator guardian based on YOLO for insulator fault detection. IEEJ Trans. Electr. Electron. Eng., 2025, 20(4): 537-547.

[39] Wang J., Cheng L.. An improved YOLOv8-XGBoost load rapid identification method based on multi-feature fusion. Int. J. Electr. Power Energy Syst., 2025, 166. ArticleID: 110573

[40] Zhang Q., Zhang J., Li Y., Zhu C., Wang G.. IL-YOLO: an efficient detection algorithm for insulator defects in complex backgrounds of transmission lines. IEEE Access, 2024, 12: 14532-14546.

[41] Li Y., Feng D., Zhang Q., Li S.. HRD-YOLOX based insulator identification and defect detection method for transmission lines. IEEE Access, 2024, 12: 22649-22661.

[42] Liu C., Wei S., Zhong S., Yu F.. YOLO-PowerLite: a lightweight YOLO model for transmission line abnormal target detection. IEEE Access, 2024, 12: 105004-105015.

[43] Hien A.T., Tran A.D., Viet D.C., Thuy Q.D.T., Huu Q.N.. Insulator defect detection based on feature pyramid network and diffusion model. Signal Image Video Process., 2025, 19(5): 1-11.

[44] Wang T., Zhang N., Zhang W., Yang W., Zhang W.. Smart grid insulator detection network improved based on YOLOv8. IEEE Latin Am. Trans., 2025, 23(2): 125-134.

[45] Ji Y., Ma T., Shen H., Feng H., Zhang Z., Li D., He Y.. Transmission line defect detection algorithm based on improved YOLOv12. Electronics, 2025, 14(12): 2432.

[46] Stefenon S.F., Seman L.O., Singh G., Yow K.-C.. Enhanced insulator fault detection using optimized ensemble of deep learning models based on weighted boxes fusion. Int. J. Electr. Power Energy Syst., 2025, 168. ArticleID: 110682

[47] Ye R., Boukerche A., Yu X.-S., Zhang C., Yan B., Zhou X.-J.. Data augmentation method for insulators based on cycle GAN. J. Electron. Sci. Technol., 2024, 22(2. ArticleID: 100250

[48] Zhang Z.-D., Zhang B., Lan Z.-C., Liu H.-C., Li D.-Y., Pei L., Yu W.-X.. FINet: an insulator dataset and detection benchmark based on synthetic fog and improved YOLOv5. IEEE Trans. Instrum. Meas., 2022, 71: 1-8.

[49] Sampedro C., Rodriguez-Vazquez J., Rodriguez-Ramos A., Carrio A., Campoy P.. Deep learning-based system for automatic recognition and diagnosis of electrical insulator strings. IEEE Access, 2019, 7: 101283-101308.

[50] Chang W., Yang G., Yu J., Liang Z.. Real-time segmentation of various insulators using generative adversarial networks. IET Comput. Vis., 2018, 12(5): 596-602.

[51] Chang W., Yang G., Wu Z., Liang Z.. Learning insulators segmentation from synthetic samples. International Joint Conference on Neural Networks (IJCNN), 2018. Rio de Janeiro, IEEE: 1-7. 1

[52] Gao Z., Yang G., Li E., Shen T., Wang Z., Tian Y., Wang H., Liang Z.. Insulator segmentation for power line inspection based on modified conditional generative adversarial network. J. Sens., 2019, 2019(1. ArticleID: 4245329

[53] Kim D., Majumder S., Xie L.. Line-post insulator fault classification model using deep convolutional gan-based synthetic images. 2023 North American Power Symposium (NAPS), 2023. Asheville, IEEE: 1-6. 1

[54] Chen Z., Wang H., Mao D., Yan Y., Rao H., Wang J.. Insulator defect detection in power transmission and transformation based on the diffusion model. 2023 IEEE 7th Conference on Energy Internet and Energy System Integration (EI2), 2023. Hangzhou, IEEE: 5248-5252.

[55] Li D., Lu Y., Gao Q., Li X., Yu X., Song Y.. LiteYOLO-ID: a lightweight object detection network for insulator defect detection. IEEE Trans. Instrum. Meas., 2024, 73: 1-12.

[56] Zhou Z., Zhang C., Xie M., Cao B.. Classification method of composite insulator surface image based on GAN and CNN. IEEE Trans. Dielectr. Electr. Insul., 2024, 31(4): 2242-2251.

[57] Liu Y., Huang X.. Efficient cross-modality insulator augmentation for multi-domain insulator defect detection in UAV images. Sensors, 2024, 24(2): 428.

[58] Yang Y., Yang S., Li C., Wang Y., Pi X., Lu Y., Wu R.. Insulator defect detection under extreme weather based on synthetic weather algorithm and improved YOLOv7. High Volt., 2024, 10(1): 69-77.

[59] Jiang D., Cao Y., Yang Q.. Weakly-supervised learning based automatic augmentation of aerial insulator images. Expert Syst. Appl., 2024, 242. ArticleID: 122739

[60] Zhang L., Wang L., Yan Z., Jia Z., Wang H., Tang X.. Star generative adversarial VGG network-based sample augmentation for insulator defect detection. Int. J. Comput. Intell. Syst., 2024, 17(1): 141.

[61] Ning P., Jin J., Xu Y., Kong C., Zhang C., Tang D., Huang J., Xu Z., Li T.. Enhanced detection of glass insulator defects using improved generative modeling and faster RCNN. Proc. CIRP, 2024, 129: 31-36.

[62] Akella R., Gunturi S.K., Sarkar D.. Enhancing power line insulator health monitoring with a hybrid generative adversarial network and YOLO3 solution. Tsinghua Sci. Technol., 2024, 29(6): 1796-1809.

[63] Liu B., Jiang W.. LA-YOLO: bidirectional adaptive feature fusion approach for small object detection of insulator self-explosion defects. IEEE Trans. Power Deliv., 2024, 39(6): 3387-3397.

[64] Terven J., Córdova-Esparza D.-M., Romero-González J.-A.. A comprehensive review of YOLO architectures in computer vision: from YOLOV1 to YOLOV8 and YOLO-NAS. Mach. Learn. Knowl. Extr., 2023, 5(4): 1680-1716.

[65] Stefenon S.F., Cristoforetti M., Cimatti A.. Towards automatic digitalization of railway engineering schematics. AIxIA 2023 – Advances in Artificial Intelligence, 2023. Rome, Springer: 453-466. 22

[66] Liu Y., Luo P.. YOLO-TS: a lightweight YOLO model for traffic sign detection. IEEE Access, 2024, 12: 169013-169023.

[67] Huang X., Zhu J., Huo Y.. SSA-YOLO: an improved YOLO for hot-rolled strip steel surface defect detection. IEEE Trans. Instrum. Meas., 2024, 73: 1-17.

[68] Stefenon S.F., Cristoforetti M., Cimatti A.. Automatic digitalization of railway interlocking systems engineering drawings based on hybrid machine learning methods. Expert Syst. Appl., 2025, 281. ArticleID: 127532

[69] Xi Y., Zhou K., Meng L.-W., Chen B., Chen H.-M., Zhang J.-Y.. Transmission line insulator defect detection based on Swin transformer and context. Mach. Intell. Res., 2023, 20(5): 729-740.

[70] Pinasthika K., Laksono B.S.P., Irsal R.B.P., Shabiyya S., Yudistira N.. Sparseswin: Swin transformer with sparse transformer block. Neurocomputing, 2024, 580. ArticleID: 127433

[71] Geng H., Chen T., Ma Y., Zhu B., Yu B.. Ptpt: physical design tool parameter tuning via multi-objective Bayesian optimization. IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst., 2023, 42(1): 178-189.

[72] Nayak J., Naik B., Dash P.B., Vimal S., Kadry S.. Hybrid Bayesian optimization hypertuned catboost approach for malicious access and anomaly detection in IoT nomalyframework. Sustain. Comput. Inform. Syst., 2022, 36. ArticleID: 100805

[73] Wang X., Jin Y., Schmitt S., Olhofer M.. Recent advances in Bayesian optimization. ACM Comput. Surv., 2023, 55(13s): 1-36.

[74] Guo R., Yang B., Guo Y., Li H., Li Z., Zhou B., Hong B., Wang F.. Machine learning-based prediction of outdoor thermal comfort: combining Bayesian optimization and the SHAP model. Build. Environ., 2024, 254. ArticleID: 111301

[75] Kyem B.A., Asamoah J.K., Huang Y., Aboah A.. Weather-adaptive synthetic data generation for enhanced power line inspection using StarGAN. IEEE Access, 2024, 12: 193882-193901.

[76] Dewi C.. Generative adversarial network for synthetic image generation method: review, analysis, and perspective. Appl. Gen. AI, 2024, 1: 91-116.

[77] She L., Fan Y., Xu M., Wang J., Xue J., Ou J.. Insulator breakage detection utilizing a convolutional neural network ensemble implemented with small sample data augmentation and transfer learning. IEEE Trans. Power Deliv., 2022, 37(4): 2787-2796.

[78] Croitoru F.-A., Hondru V., Ionescu R.T., Shah M.. Diffusion models in vision: a survey. IEEE Trans. Pattern Anal. Mach. Intell., 2023, 45(9): 10850-10869.

[79] Yang L., Zhang Z., Song Y., Hong S., Xu R., Zhao Y., Zhang W., Cui B., Yang M.-H.. Diffusion models: a comprehensive survey of methods and applications. ACM Comput. Surv., 2023, 56(4): 1-39.

[80] Xu Z., Wang S., Zhang X.-L., He G.. Optimal sensor placement for ensemble-based data assimilation using gradient-weighted class activation mapping. J. Comput. Phys., 2024, 514. ArticleID: 113224

[81] Lundberg S.M., Lee S.-I.. Guyon I., Luxburg U.V., Bengio S., Wallach H., Fergus R., Vishwanathan S., Garnett R.. A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 20174765-477431

[82] Kwak J., Lee Y., Choi M., Lee S.. Deep learning based approaches to enhance energy efficiency in autonomous driving systems. Energy, 2024, 307. ArticleID: 132625

[83] Kim Y., Kim Y.. Explainable heat-related mortality with random forest and Shapley additive explanations (SHAP) models. Sustain. Cities Soc., 2022, 79. ArticleID: 103677

[84] Sun K.H., Huh H., Tama B.A., Lee S.Y., Jung J.H., Lee S.. Vision-based fault diagnostics using explainable deep learning with class activation maps. IEEE Access, 2020, 8: 129169-129179.

[85] Selvaraju R.R., Cogswell M., Das A., Vedantam R., Parikh D., Batra D.. GRAD-CAM: visual explanations from deep networks via gradient-based localization. Int. J. Comput. Vis., 2020, 128: 336-359.

[86] R.L. Draelos, L. Carin, Use hirescam instead of grad-cam for faithful explanations of convolutional neural networks, arXiv preprint (2021). https://doi.org/10.48550/arXiv.2011.08891. arXiv:2011.08891

[87] Wang H., Wang Z., Du M., Yang F., Zhang Z., Ding X.H., Mardziel P., Hu X.L.. Score-cam: score-weighted visual explanations for convolutional neural networks. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2020. Seattle, IEEE: 24-25.

[88] Zhao W., Chen J., Liu X., Liu J.. Enhancing object detection through global collaborative learning. Auton. Intell. Syst., 2025, 5(1): 29.

[89] Li Y., Miao N., Ma L., Shuang F., Huang X.. Transformer for object detection: review and benchmark. Eng. Appl. Artif. Intell., 2023, 126. ArticleID: 107021

[90] Ou J., Wang J., Xue J., Wang J., Zhou X., She L., Fan Y.. Infrared image target detection of substation electrical equipment using an improved faster R-CNN. IEEE Trans. Power Deliv., 2023, 38(1): 387-396.

[91] Teju V., Sowmya K.V., Kandula S.R., Stan A., Stan O.P.. A hybrid retina net classifier for thermal imaging. Appl. Sci., 2023, 13(14): 8525.

[92] Miao X., Liu X., Chen J., Zhuang S., Fan J., Jiang H.. Insulator detection in aerial images for transmission line inspection using single shot multibox detector. IEEE Access, 2019, 7: 9945-9956.

[93] Vigneshwaran B., Maheswari R., Kalaivani L., Shanmuganathan V., Rho S., Kadry S., Lee M.Y.. Recognition of pollution layer location in 11 kv polymer insulators used in smart power grid using dual-input VGG convolutional neural network. Energy Rep., 2021, 7: 7878-7889.

[94] Wang S., Liu Y., Qing Y., Wang C., Lan T., Yao R.. Detection of insulator defects with improved ResNeSt and region proposal network. IEEE Access, 2020, 8: 184841-184850.

[95] Swetapadma A., Yadav A.. Transfer learning based EfficientNet method for transmission line insulator flaw detection. e-Prime - Adv. Electr. Eng. Electron. Energy, 2024, 10. ArticleID: 100873

[96] Surek G.A.S., Seman L.O., Stefenon S.F., Mariani V.C., Coelho L.d.S.. Video-based human activity recognition using deep learning approaches. Sensors, 2023, 23(14): 6384.

[97] Saad M.M., O’Reilly R., Rehmani M.H.. A survey on training challenges in generative adversarial networks for biomedical image analysis. Artif. Intell. Rev., 2024, 57(2): 19.

[98] D. Lewis, P. Kulkarni, Insulator Defect Detection. IEEE Dataport (2021). https://ieee-dataport.org/competitions/insulator-defect-detection. Accessed on March 15, 2025

Funding

Natural Sciences and Engineering Research Council of Canada (NSERC)(DG-2024-00035)

RIGHTS & PERMISSIONS

The Author(s)