{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,8]],"date-time":"2026-09-08T05:10:49Z","timestamp":1788844249803,"version":"build-2803163510"},"reference-count":45,"publisher":"MDPI AG","issue":"17","license":[{"start":{"date-parts":[[2026,9,5]],"date-time":"2026-09-05T00:00:00Z","timestamp":1788566400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Institute for Occupational Safety and Healt","award":["U60OH012685-01-00"],"award-info":[{"award-number":["U60OH012685-01-00"]}]},{"award":["U60OH012685-01-00"],"award-info":[{"award-number":["U60OH012685-01-00"]}],"id":[{"id":"https:\/\/ror.org\/0502a2655","id-type":"ROR","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000030","name":"Centers for Disease Control and Prevention","doi-asserted-by":"crossref","id":[{"id":"10.13039\/100000030","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Reliable personnel detection is critical for the safe deployment of autonomous haulage systems in underground mining, where challenging environmental conditions demand robust real-time perception. Existing research has focused primarily on fine-tuned convolutional detectors, while systematic comparisons with zero-shot vision-language models remain limited. This study presents a cross-paradigm benchmark comparing four zero-shot vision-language models (YOLO-World, Grounding DINO, OWL-ViT, and OWLv2) with four fine-tuned YOLO detectors (YOLOv8s, YOLOv9s, YOLO11s, and YOLO26s) using 31,396 real-world underground coal mine images. Detection performance was evaluated using precision, recall, F1-score, average precision, inference speed, and condition- and target scale-specific recall. The experimental results show that the fine-tuned detectors achieved F1-scores of 0.8449\u20130.8662 and AP50 values of 0.8791\u20130.9135, with YOLO26s achieving the strongest overall performance. In comparison, the zero-shot models achieved F1-scores of 0.3186\u20130.5484 and AP50 values of 0.2484\u20130.5108, with Grounding DINO performing best among the zero-shot models. Fine-tuned detectors also maintained substantially higher recall for occluded personnel and small apparent targets. These findings demonstrate a substantial performance advantage for domain-specific fine-tuning over the evaluated zero-shot approaches and establish a controlled cross-paradigm benchmark for comparing detection paradigms for underground personnel perception.<\/jats:p>","DOI":"10.3390\/s26175646","type":"journal-article","created":{"date-parts":[[2026,9,7]],"date-time":"2026-09-07T12:12:25Z","timestamp":1788783145000},"page":"5646","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Benchmarking Zero-Shot Open-Vocabulary and Fine-Tuned Object Detectors for Underground Mine Personnel Detection"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-2648-1916","authenticated-orcid":false,"given":"Ellen","family":"Essien","sequence":"first","affiliation":[{"name":"Department of Mining and Explosives Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USA"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"conceptualization","vocabulary":"credit"},{"role":"methodology","vocabulary":"credit"},{"role":"software","vocabulary":"credit"},{"role":"data-curation","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"},{"role":"formal-analysis","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8300-6824","authenticated-orcid":false,"given":"Samuel","family":"Frimpong","sequence":"additional","affiliation":[{"name":"Department of Mining and Explosives Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USA"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"resources","vocabulary":"credit"},{"role":"funding-acquisition","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"},{"role":"supervision","vocabulary":"credit"},{"role":"project-administration","vocabulary":"credit"}]}],"member":"1968","published-online":{"date-parts":[[2026,9,5]]},"reference":[{"key":"ref_1","unstructured":"(2025, April 23). Caterpillar, \u201cMilestone: Caterpillar Surpasses 5 Billion Tonnes of Material Autonomously Hauled,\u201d Caterpillar. Available online: https:\/\/www.cat.com\/en_US\/news\/machine-press-releases\/milestone-caterpillar-surpasses-5-billion-tonnes-of-material-autonomously-hauled.html."},{"key":"ref_2","unstructured":"(2026, July 09). Komatsu Becomes First OEM to Commission 1000 Ultra-Class Autonomous Haul Trucks. Available online: https:\/\/www.komatsu.com\/en-us\/newsroom\/2026\/komatsu-becomes-first-oem-to-commission-1000-ultra-class-autonomous-haul-trucks."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Ni, Y., Huo, J., Hou, Y., Wang, J., and Guo, P. (2024). Detection of Underground Dangerous Area Based on Improving YOLOV8. Electronics, 13.","DOI":"10.3390\/electronics13030623"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Xu, X., Li, L., and Zhang, Q. (2023). Challenges of Autonomous Navigation and Perception Technology for Unmanned Special Vehicles in Underground Mine. Proceedings of the 2023 6th International Symposium on Autonomous Systems (ISAS), IEEE.","DOI":"10.1109\/ISAS59543.2023.10164322"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"4019","DOI":"10.1007\/s40747-024-01387-2","article-title":"A low-light image enhancement method for personnel safety monitoring in underground coal mines","volume":"10","author":"Yang","year":"2024","journal-title":"Complex Intell. Syst."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Teng, S., Li, X., Li, Y., Li, L., and Chen, L. (2023). Scenarios Engineering Driven Autonomous Transportation in Open-Pit Mines. Proceedings of the 2023 IEEE 3rd International Conference on Digital Twins and Parallel Intelligence (DTPI), IEEE.","DOI":"10.1109\/DTPI59677.2023.10365481"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Pei, S., and Yang, J. (2023). Multi-agent Deep Reinforcement Learning to Improve Efficiency of Autonomous Mining Transportation. SSRN.","DOI":"10.2139\/ssrn.4604945"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Mao, M., and Hong, M. (2025). YOLO Object Detection for Real-Time Fabric Defect Inspection in the Textile Industry: A Review of YOLOv1 to YOLOv11. Sensors, 25.","DOI":"10.3390\/s25072270"},{"key":"ref_9","unstructured":"Somua-Gyimah, G., Frimpong, S., and Gbadam, E. (2019). A computer vision system for terrain recognition and object detection tasks in mining and construction environments. Proceedings of the SME Annual Conference & Expo and CMA 121st National Western Mining Conference, Society for Mining, Metallurgy & Exploration. Available online: https:\/\/www.researchgate.net\/publication\/330130008."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Zhang, Y., and Zhou, Y. (2021). YOLOv5 Based Pedestrian Safety Detection in Underground Coal Mines. Proceedings of the 2021 IEEE International Conference on Robotics and Biomimetics (ROBIO), IEEE.","DOI":"10.1109\/ROBIO54168.2021.9739594"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Chen, Y., Meng, X., Yang, C., and Wang, Y. (2026). LLE-YOLO: Adaptive Low-Light-Enhanced and Degradation-Aware Multi-Scale Attention Network for Miner Detection in Underground Coal Mines. Appl. Sci., 16.","DOI":"10.3390\/app16104983"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Long, M., Schafrik, S., Kolapo, P., Agioutantis, Z., and Sottile, J. (2024). Equipment and Operations Automation in Mining: A Review. Machines, 12.","DOI":"10.3390\/machines12100713"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"295","DOI":"10.17159\/2411-9717\/1065\/2021","article-title":"Navigation and positioning technology in underground coal mines and tunnels: A review","volume":"121","author":"Cui","year":"2021","journal-title":"J. S. Afr. Inst. Min. Metall."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Imam, M., Ba\u00efna, K., Tabii, Y., Ressami, E.M., Adlaoui, Y., Benzakour, I., and Abdelwahed, E.h. (2023). The Future of Mine Safety: A Comprehensive Review of Anti-Collision Systems Based on Computer Vision in Underground Mines. Sensors, 23.","DOI":"10.3390\/s23094294"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Liu, S., Zeng, Z.Y., Ren, T.H., Li, F., Zhang, H., Yang, J., Jiang, Q., Li, C.Y., Yang, J.W., and Su, H. (2024). Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection. Proceedings of the Computer Vision\u2013ECCV 2024, Springer Nature.","DOI":"10.1007\/978-3-031-72970-6_3"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Minderer, M., Gritsenko, A., Stone, A., Neumann, M., Weissenborn, D., Dosovitskiy, A., Mahendran, A., Arnab, A., Dehghani, M., and Shen, Z.R. (2022). Simple Open-Vocabulary Object Detection. Proceedings of the Computer Vision\u2013ECCV 2022, Springer.","DOI":"10.1007\/978-3-031-20080-9_42"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Cheng, T., Song, L., Ge, Y., Liu, W., Wang, X., and Shan, Y. (2024). YOLO-World: Real-Time Open-Vocabulary Object Detection. Proceedings of the 2024 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 16\u201322 June 2024, IEEE.","DOI":"10.1109\/CVPR52733.2024.01599"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"864","DOI":"10.1016\/j.psep.2024.04.137","article-title":"Analysis of the causes and safety countermeasures of coal mine accidents: A case study of coal mine accidents in China from 2018 to 2022","volume":"187","author":"Tian","year":"2024","journal-title":"Process Saf. Environ. Prot."},{"key":"ref_19","first-page":"78","article-title":"Coal Mine Safety Production Situation and Management Strategy","volume":"14","author":"Niu","year":"2014","journal-title":"Manag. Eng."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2016). You only look once: Unified, real-time object detection. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE Computer Society.","DOI":"10.1109\/CVPR.2016.91"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014). Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA, 23\u201328 June 2014, IEEE.","DOI":"10.1109\/CVPR.2014.81"},{"key":"ref_22","unstructured":"Jocher, G., Chaurasia, A., and Qiu, J. (2023). Ultralytics YOLOv8, Ultralytics. Available online: https:\/\/github.com\/ultralytics\/ultralytics."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Wang, C.Y., Yeh, I.H., and Liao, H.Y.M. (2024). YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information. Proceedings of the Computer Vision\u2013ECCV 2024, Springer.","DOI":"10.1007\/978-3-031-72751-1_1"},{"key":"ref_24","unstructured":"(2025, December 07). Ultralytics, \u201cUltralytics YOLO11,\u201d Ultralytics Docs. Available online: https:\/\/docs.ultralytics.com\/models\/yolo11\/."},{"key":"ref_25","unstructured":"Jocher, G., Qiu, J., Liu, M., Lyu, S., Akyon, F.C., and Kalfaoglu, M.E. (2026). Ultralytics YOLO26: Unified Real-Time End-to-End Vision Models. arXiv."},{"key":"ref_26","unstructured":"Sapkota, R., and Karkee, M. (2025). Ultralytics YOLO Evolution: An Overview of YOLO26, YOLO11, YOLOv8 and YOLOv5 Object Detectors for Computer Vision and Pattern Recognition. arXiv."},{"key":"ref_27","first-page":"013017","article-title":"Improved YOLOv4 network using infrared images for personnel detection in coal mines","volume":"31","author":"Li","year":"2022","journal-title":"J. Electron. Imaging"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Zhou, F., Zou, J., Xue, R., Yu, M., Wang, X., Xue, W., and Yao, S. (2025). Enhancing Object Detection in Underground Mines: UCM-Net and Self-Supervised Pre-Training. Sensors, 25.","DOI":"10.3390\/s25072103"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"e70051","DOI":"10.1002\/cav.70051","article-title":"Yolov8-HAC: Safety Helmet Detection Model for Complex Underground Coal Mine Scene","volume":"36","author":"Liu","year":"2025","journal-title":"Comput. Animat. Virtual Worlds"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"116624","DOI":"10.1016\/j.measurement.2024.116624","article-title":"Research on mine personnel target detection method based on improved YOLOv8","volume":"245","author":"Jin","year":"2025","journal-title":"Measurement"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Zhang, L., Tao, H., Sun, Z., and Yi, W. (2025). Multi-Strategy Enhancement of YOLOv8n Monitoring Method for Personnel and Vehicles in Mine Air Door Scenarios. Sensors, 25.","DOI":"10.3390\/s25103128"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Zhang, L., Sun, Z., Tao, H., Wang, M., and Yi, W. (2024). Research on Mine-Personnel Helmet Detection Based on Multi-Strategy-Improved YOLOv11. Sensors, 25.","DOI":"10.3390\/s25010170"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Li, Y., Yan, H., Li, D., and Wang, H. (2024). Robust Miner Detection in Challenging Underground Environments: An Improved YOLOv11 Approach. Appl. Sci., 14.","DOI":"10.3390\/app142411700"},{"key":"ref_34","unstructured":"Radford, A., Kim, J.W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., and Clark, J. (2021). Learning transferable visual models from natural language supervision. Proceedings of the 38th International Conference on Machine Learning, PMLR. Available online: https:\/\/proceedings.mlr.press\/v139\/radford21a.html."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Minderer, M., Gritsenko, A., and Houlsby, N. (2023). Scaling open-vocabulary object detection. Proceedings of the Advances in Neural Information Processing Systems 36 (NeurIPS 2023), Neural Information Processing Systems Foundation.","DOI":"10.52202\/075280-3191"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Choi, L., and Greer, R. (2024). Evaluating Cascaded Methods of Vision-Language Models for Zero-Shot Detection and Association of Hardhats for Increased Construction Safety. Proceedings of the 2024 IEEE MIT Undergraduate Research Technology Conference (URTC), IEEE.","DOI":"10.1109\/URTC65039.2024.10937518"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Wu, J., Wu, S.C., Ma, Y.S., Yu, G.Y., Xu, H.Y., Zheng, L.F., and Duan, J.L. (2025). MonitorVLM: A Vision Language Framework for Safety Violation Detection in Mining Operations. arXiv.","DOI":"10.21203\/rs.3.rs-10403842\/v1"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Abdullah, R., Rawat, Y., and Vyas, S. (2025). ISafetyBench: A video-language benchmark for safety in industrial environment. Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV) Workshops, IEEE.","DOI":"10.1109\/ICCVW69036.2025.00154"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Chhipa, P.C., De, K., Chippa, M.S., Saini, R., and Liwicki, M. (2024). Open-Vocabulary Object Detectors: Robustness Challenges Under Distribution Shifts. Computer Vision\u2013ECCV 2024 Workshops, Springer.","DOI":"10.1007\/978-3-031-91672-4_5"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"303","DOI":"10.1007\/s11263-009-0275-4","article-title":"The Pascal Visual Object Classes (VOC) Challenge","volume":"88","author":"Everingham","year":"2010","journal-title":"Int. J. Comput. Vis."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Doll\u00e1r, P., and Lawrence Zitnick, C. (2014). Microsoft COCO: Common Objects in Context. Computer Vision\u2013ECCV 2014, Springer.","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Popov, M., Robicheaux, P., Madan, A., Robinson, I., Nelson, J., Ramanan, D., and Peri, N. (2025). Roboflow100-VL: A Multi-Domain Object Detection Benchmark for Vision-Language Models. Advances in Neural Information Processing Systems 38 (NeurIPS 2025), Neural Information Processing Systems Foundation.","DOI":"10.52202\/085713-0363"},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Madan, A., Peri, N., Kong, S., and Ramanan, D. (2024). Revisiting Few-Shot Object Detection with Vision-Language Models. Advances in Neural Information Processing Systems 37 (NeurIPS 2024), Neural Information Processing Systems Foundation.","DOI":"10.52202\/079017-0617"},{"key":"ref_44","unstructured":"Sapkota, R., Roumeliotis, K.I., Karkee, M., and Tselikas, N.D. (2025). Generalization vs. Specialization: Evaluating Segment Anything Model (SAM3) Zero-Shot Segmentation Against Fine-Tuned YOLO Detectors. arXiv."},{"key":"ref_45","first-page":"300","article-title":"A dataset of drilling site object detection in underground coal mines","volume":"9","author":"Zhou","year":"2024","journal-title":"China Sci. Data"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/17\/5646\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,8]],"date-time":"2026-09-08T04:28:44Z","timestamp":1788841724000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/17\/5646"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,5]]},"references-count":45,"journal-issue":{"issue":"17","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["s26175646"],"URL":"https:\/\/doi.org\/10.3390\/s26175646","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,5]]}}}