{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T17:59:57Z","timestamp":1790359197671,"version":"4.1.0"},"reference-count":45,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T00:00:00Z","timestamp":1790294400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Continuous, reliable monitoring of geographically dispersed oil and gas production facilities is essential for accident prevention, asset integrity, and regulatory compliance, yet ground-based inspection is labour-intensive and temporally sparse. Unmanned aerial vehicles (UAVs) offer on-demand aerial coverage, but automated interpretation of production-site imagery is hindered by extreme target-scale variation, irregular equipment geometries, and the limited compute available on airborne platforms. This paper presents an integrated lightweight detection-and-tracking framework tailored to oil and gas field surveillance. The detector augments YOLOv8s with deformable C2f (DCNv2) modules for geometry-adaptive feature extraction, a progressive feature pyramid (AFPN) that suppresses cross-scale semantic conflict, a parameter-shared detail-enhanced head (LSDECD-Head), a Focaler-GIoU regression loss, and LAMP structured pruning for embedded deployment. For dynamic situational awareness, an improved ByteTrack introduces a spatial\u2013appearance similarity matrix encoding operational-state cues and an acceleration-aware Kalman correction. On VisDrone2019, UAVDT, and a self-constructed oil-and-gas dataset (OGF-UAV), the method attains 44.1%, 57.6%, and 56.3% mAP@0.5 with only 7.8 M parameters at 78 FPS, and reduces tracking identity switches by up to 43.2%. The pruned model runs at 18.3 FPS on an NVIDIA Jetson Orin NX 16 GB under TensorRT FP16, confirming practicality for real-time field deployment.<\/jats:p>","DOI":"10.1371\/journal.pone.0358757","type":"journal-article","created":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T17:25:59Z","timestamp":1790357159000},"page":"e0358757","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Lightweight UAV-based object detection and tracking for intelligent oil and gas field safety monitoring"],"prefix":"10.1371","volume":"21","author":[{"given":"Jianhua","family":"Gong","sequence":"first","affiliation":[{"name":"Digital Intelligent Technology Subsidiary of Southwest Oil and Gas Field","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-8942-2733","authenticated-orcid":true,"given":"Yifu","family":"Wang","sequence":"additional","affiliation":[{"name":"Digital Intelligent Technology Subsidiary of Southwest Oil and Gas Field","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jun","family":"Zhang","sequence":"additional","affiliation":[{"name":"Digital Intelligent Technology Subsidiary of Southwest Oil and Gas Field","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Defei","family":"Du","sequence":"additional","affiliation":[{"name":"Digital Intelligent Technology Subsidiary of Southwest Oil and Gas Field","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yang","family":"Huang","sequence":"additional","affiliation":[{"name":"Digital Intelligent Technology Subsidiary of Southwest Oil and Gas Field","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,25]]},"reference":[{"issue":"1","key":"pone.0358757.ref001","doi-asserted-by":"crossref","first-page":"149","DOI":"10.3390\/rs16010149","article-title":"A survey of object detection for UAVs based on deep learning","volume":"16","author":"G Tang","year":"2023","journal-title":"Remote Sens"},{"issue":"5","key":"pone.0358757.ref002","doi-asserted-by":"crossref","first-page":"4295","DOI":"10.1007\/s10462-022-10281-7","article-title":"Swarm intelligence algorithms for multiple unmanned aerial vehicles collaboration: a comprehensive review","volume":"56","author":"J Tang","year":"2022","journal-title":"Artif Intell Rev"},{"key":"pone.0358757.ref003","doi-asserted-by":"crossref","first-page":"110936","DOI":"10.1016\/j.engappai.2025.110936","article-title":"A small object detection algorithm for mine environment","volume":"153","author":"D Liu","year":"2025","journal-title":"Eng Appl Artif Intell"},{"key":"pone.0358757.ref004","doi-asserted-by":"crossref","first-page":"128341","DOI":"10.1016\/j.eswa.2025.128341","article-title":"Adaptive feature fusion and task-dynamic alignment for real-time object detection on edge devices","volume":"290","author":"Y Ding","year":"2025","journal-title":"Expert Syst Appl"},{"issue":"23","key":"pone.0358757.ref005","doi-asserted-by":"crossref","first-page":"40017","DOI":"10.1109\/JSEN.2024.3478810","article-title":"Enhancing smart agriculture with lightweight object detection: MobileNetv3-YOLOv4 and adaptive width multipliers","volume":"24","author":"I-A Lin","year":"2024","journal-title":"IEEE Sensors J"},{"key":"pone.0358757.ref006","doi-asserted-by":"crossref","first-page":"103448","DOI":"10.1016\/j.artint.2020.103448","article-title":"Multiple object tracking: a literature review","volume":"293","author":"W Luo","year":"2021","journal-title":"Artif Intell"},{"key":"pone.0358757.ref007","first-page":"1","author":"Y Zhang","year":"2022","journal-title":"Bytetrack: Multi-object tracking by associating every detection box[C]\/\/European conference on computer vision"},{"key":"pone.0358757.ref008","first-page":"9686","author":"J Cao","year":"2023","journal-title":"Observation-centric sort: rethinking sort for robust multi-object tracking[C]\/\/Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition"},{"key":"pone.0358757.ref009","unstructured":"Jocher G, Chaurasia A, Qiu J. Ultralytics YOLOv8. GitHub 2023. Available from: https:\/\/github.com\/ultralytics\/ultralytics"},{"key":"pone.0358757.ref010","doi-asserted-by":"crossref","first-page":"99","DOI":"10.1016\/j.neucom.2022.06.116","article-title":"RMDC: rotation-mask deformable convolution for object detection in top-view fisheye cameras","volume":"504","author":"X Wei","year":"2022","journal-title":"Neurocomputing"},{"key":"pone.0358757.ref011","doi-asserted-by":"crossref","first-page":"439","DOI":"10.1016\/j.neunet.2022.08.029","article-title":"Attentional feature pyramid network for small object detection","volume":"155","author":"K Min","year":"2022","journal-title":"Neural Netw"},{"key":"pone.0358757.ref012","doi-asserted-by":"crossref","first-page":"4354","DOI":"10.1109\/LSP.2025.3628795","article-title":"SDWS-YOLO: a lightweight small object detection model for aerial remote sensing images","volume":"32","author":"J Bi","year":"2025","journal-title":"IEEE Signal Process Lett"},{"key":"pone.0358757.ref013","doi-asserted-by":"crossref","first-page":"1029","DOI":"10.1016\/j.neucom.2022.06.018","article-title":"Absolute size IoU loss for the bounding box regression of the object detection","volume":"500","author":"D Tian","year":"2022","journal-title":"Neurocomputing"},{"key":"pone.0358757.ref014","article-title":"Efficient distortion-minimized layerwise pruning[J","author":"K Xu","year":"2025","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"pone.0358757.ref015","doi-asserted-by":"crossref","unstructured":"Zhu X, Lyu S, Wang X, Zhao Q. TPH-YOLOv5: improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios. In: 2021 IEEE\/CVF International Conference on Computer Vision Workshops (ICCVW); 2021. pp. 2778\u201388. https:\/\/doi.org\/10.1109\/iccvw54120.2021.00312","DOI":"10.1109\/ICCVW54120.2021.00312"},{"key":"pone.0358757.ref016","doi-asserted-by":"crossref","first-page":"116484","DOI":"10.1016\/j.image.2021.116484","article-title":"Multiview vision-based human crowd localization for UAV fleet flight safety","volume":"99","author":"E Kakaletsis","year":"2021","journal-title":"Signal Process Image Commun"},{"issue":"10","key":"pone.0358757.ref017","doi-asserted-by":"crossref","first-page":"13863","DOI":"10.1109\/TITS.2024.3386928","article-title":"YOLC: you only look clusters for tiny object detection in aerial images","volume":"25","author":"C Liu","year":"2024","journal-title":"IEEE Trans Intell Transport Syst"},{"issue":"12","key":"pone.0358757.ref018","doi-asserted-by":"crossref","first-page":"3299","DOI":"10.1007\/s11263-021-01523-2","article-title":"NAS-FCOS: efficient search for object detection architectures","volume":"129","author":"N Wang","year":"2021","journal-title":"Int J Comput Vis"},{"issue":"2","key":"pone.0358757.ref019","doi-asserted-by":"crossref","first-page":"229","DOI":"10.3390\/rs18020229","article-title":"Oil and gas facility detection in high-resolution remote sensing images based on oriented R-CNN","volume":"18","author":"Y Qian","year":"2026","journal-title":"Remote Sens"},{"issue":"3","key":"pone.0358757.ref020","doi-asserted-by":"crossref","first-page":"370","DOI":"10.3390\/agronomy16030370","article-title":"YOLOv11-IMP: anchor-free multiscale detection model for accurate grape yield estimation in precision viticulture","volume":"16","author":"S Zheng","year":"2026","journal-title":"Agronomy"},{"issue":"3","key":"pone.0358757.ref021","doi-asserted-by":"crossref","first-page":"1031","DOI":"10.32604\/csse.2022.023221","article-title":"Planetscope nanosatellites image classification using machine learning","volume":"42","author":"M Anul Haq","year":"2022","journal-title":"Comput Syst Sci Eng"},{"issue":"2","key":"pone.0358757.ref022","doi-asserted-by":"crossref","first-page":"837","DOI":"10.32604\/csse.2022.023016","article-title":"CNN Based automated weed detection system using UAV imagery","volume":"42","author":"M Anul Haq","year":"2022","journal-title":"Comput Syst Sci Eng"},{"issue":"3","key":"pone.0358757.ref023","doi-asserted-by":"crossref","first-page":"601","DOI":"10.1007\/s12524-020-01231-3","article-title":"Deep learning based supervised image classification using UAV images for forest areas classification","volume":"49","author":"MA Haq","year":"2020","journal-title":"J Indian Soc Remote Sens"},{"key":"pone.0358757.ref024","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/LGRS.2024.3362890","article-title":"Deformable convolution-guided multiscale feature learning and fusion for UAV object detection","volume":"21","author":"Y Shi","year":"2024","journal-title":"IEEE Geosci Remote Sensing Lett"},{"key":"pone.0358757.ref025","first-page":"7464","author":"C Y Wang","year":"2023","journal-title":"YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]\/\/Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition"},{"key":"pone.0358757.ref026","doi-asserted-by":"crossref","first-page":"108159","DOI":"10.1016\/j.patcog.2021.108159","article-title":"Delving deep into spatial pooling for squeeze-and-excitation networks","volume":"121","author":"X Jin","year":"2022","journal-title":"Pattern Recogn"},{"key":"pone.0358757.ref027","doi-asserted-by":"crossref","first-page":"108418","DOI":"10.1016\/j.patcog.2021.108418","article-title":"Sparse attention block: aggregating contextual information for object detection","volume":"124","author":"C Chen","year":"2022","journal-title":"Pattern Recogn"},{"key":"pone.0358757.ref028","first-page":"11534","author":"Q Wang","year":"2020","journal-title":"ECA-Net: efficient channel attention for deep convolutional neural networks[C]\/\/Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition"},{"key":"pone.0358757.ref029","doi-asserted-by":"crossref","first-page":"4341","DOI":"10.1109\/TIP.2023.3297408","article-title":"Centralized feature pyramid for object detection","volume":"32","author":"Y Quan","year":"2023","journal-title":"IEEE Trans Image Process"},{"issue":"1","key":"pone.0358757.ref030","doi-asserted-by":"crossref","first-page":"595","DOI":"10.1109\/TNNLS.2023.3336563","article-title":"Learning to aggregate multi-scale context for instance segmentation in remote sensing images","volume":"36","author":"Y Liu","year":"2025","journal-title":"IEEE Trans Neural Netw Learn Syst"},{"issue":"3","key":"pone.0358757.ref031","doi-asserted-by":"crossref","first-page":"516","DOI":"10.3390\/rs14030516","article-title":"An adaptive attention fusion mechanism convolutional network for object detection in remote sensing images","volume":"14","author":"Y Ye","year":"2022","journal-title":"Remote Sens"},{"issue":"5","key":"pone.0358757.ref032","doi-asserted-by":"crossref","first-page":"100466","DOI":"10.1016\/j.jpse.2026.100466","article-title":"FPE\u2011YOLO: real\u2011time small\u2011object detection for third\u2011party damage in oil and gas pipeline integrity monitoring","volume":"6","author":"S Dong","year":"2026","journal-title":"J Pipeline Sci Eng"},{"key":"pone.0358757.ref033","first-page":"21002","article-title":"Generalized focal loss: learning qualified and distributed bounding boxes for dense object detection[J].","volume":"33","author":"X Li","year":"2020","journal-title":"Adv Neural Inf Process Syst"},{"key":"pone.0358757.ref034","doi-asserted-by":"crossref","unstructured":"Feng C, Zhong Y, Gao Y, Scott MR, Huang W. TOOD: task-aligned one-stage object detection. In: 2021 IEEE\/CVF International Conference on Computer Vision (ICCV); 2021. pp. 3490\u20139. https:\/\/doi.org\/10.1109\/iccv48922.2021.00349","DOI":"10.1109\/ICCV48922.2021.00349"},{"issue":"9","key":"pone.0358757.ref035","doi-asserted-by":"crossref","DOI":"10.1007\/s10462-025-11253-3","article-title":"YOLO advances to its genesis: a decadal and comprehensive review of the You Only Look Once (YOLO) series","volume":"58","author":"R Sapkota","year":"2025","journal-title":"Artif Intell Rev"},{"key":"pone.0358757.ref036","doi-asserted-by":"crossref","first-page":"671","DOI":"10.1109\/TIP.2023.3348697","article-title":"Probabilistic intersection-over-union for training and evaluation of oriented object detectors","volume":"33","author":"J Murrugarra-Llerena","year":"2024","journal-title":"IEEE Trans Image Process"},{"issue":"07","key":"pone.0358757.ref037","doi-asserted-by":"crossref","first-page":"12993","DOI":"10.1609\/aaai.v34i07.6999","article-title":"Distance-IoU loss: faster and better learning for bounding box regression","volume":"34","author":"Z Zheng","year":"2020","journal-title":"AAAI"},{"key":"pone.0358757.ref038","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TGRS.2025.3580937","article-title":"Incorporating multiscale context and task-consistent focal loss into oriented object detection","volume":"63","author":"X Qian","year":"2025","journal-title":"IEEE Trans Geosci Remote Sensing"},{"key":"pone.0358757.ref039","doi-asserted-by":"crossref","first-page":"61","DOI":"10.1016\/j.neucom.2019.11.023","article-title":"Deep learning in video multi-object tracking: A survey","volume":"381","author":"G Ciaparrone","year":"2020","journal-title":"Neurocomputing"},{"key":"pone.0358757.ref040","doi-asserted-by":"crossref","first-page":"109982","DOI":"10.1016\/j.engappai.2024.109982","article-title":"Underwater moving target detection and tracking based on enhanced you only look once and deep simple online and realtime tracking strategy","volume":"143","author":"B Sun","year":"2025","journal-title":"Eng Appl Artif Intell"},{"key":"pone.0358757.ref041","doi-asserted-by":"crossref","first-page":"1439","DOI":"10.1109\/TIP.2020.3044219","article-title":"Tracking-by-counting: using network flows on crowd density maps for tracking multiple targets","volume":"30","author":"W Ren","year":"2021","journal-title":"IEEE Trans Image Process"},{"key":"pone.0358757.ref042","doi-asserted-by":"crossref","first-page":"126772","DOI":"10.1016\/j.eswa.2025.126772","article-title":"Pedestrian multi-object tracking combining appearance and spatial characteristics","volume":"272","author":"Y Guo","year":"2025","journal-title":"Expert Syst Appl"},{"key":"pone.0358757.ref043","doi-asserted-by":"crossref","first-page":"108298","DOI":"10.1016\/j.compag.2023.108298","article-title":"Transforming unmanned pineapple picking with spatio-temporal convolutional neural networks","volume":"214","author":"F Meng","year":"2023","journal-title":"Comput Electron Agric"},{"issue":"11","key":"pone.0358757.ref044","doi-asserted-by":"crossref","first-page":"7380","DOI":"10.1109\/TPAMI.2021.3119563","article-title":"Detection and tracking meet drones challenge","volume":"44","author":"P Zhu","year":"2022","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"pone.0358757.ref045","first-page":"370","author":"D Du","year":"2018","journal-title":"The unmanned aerial vehicle benchmark: Object detection and tracking[C]\/\/Proceedings of the European conference on computer vision (ECCV"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358757","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T17:26:09Z","timestamp":1790357169000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0358757"}},"subtitle":[],"editor":[{"given":"Yang","family":"Lou","sequence":"first","affiliation":[{"name":"Hiroshima University: Hiroshima Daigaku","place":["JAPAN"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,25]]},"references-count":45,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,25]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0358757","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,25]]}}}