{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,16]],"date-time":"2026-03-16T18:37:01Z","timestamp":1773686221270,"version":"3.50.1"},"reference-count":46,"publisher":"Public Library of Science (PLoS)","issue":"3","license":[{"start":{"date-parts":[[2026,3,16]],"date-time":"2026-03-16T00:00:00Z","timestamp":1773619200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100012579","name":"Natural Science Foundation of Qinghai Province","doi-asserted-by":"publisher","award":["2023-ZJ-989Q"],"award-info":[{"award-number":["2023-ZJ-989Q"]}],"id":[{"id":"10.13039\/501100012579","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>In mixed traffic flow scenarios, multiple types of traffic participants coexist on the same roadway, posing severe challenges for object detection algorithms due to significant disparities in target scales, complex background interference, dense occlusions, and the high heterogeneity of classes. Existing CNN-based detectors are constrained by the fixed receptive fields inherent in convolution operations and are generally plagued by imbalances between positive and negative samples as well as inadequate representations of small objects, further limiting their performance in mixed traffic detection tasks. To address these issues, we propose the MTF-NET detection network, which is endowed with full-field perceptual capabilities. First, a combination of CNN and MetaFormer is employed as the backbone for feature extraction to enhance contextual modeling. Second, to mitigate the inherent dual-dimensional information loss and small-target representation bottlenecks associated with pyramid structures, we introduce a Hierarchical Implicit-Explicit Pyramid structure alongside a Multi-Kernel Dilation Fusion Network designed to counteract the information degradation brought about by pooling operations. Finally, the Dynamic Dual Detection Heads utilize a dual-branch design that facilitates end-to-end deployment while alleviating the limitations imposed by non-maximum suppression (NMS), and a hybrid strategy integrating Exponential Adaptive Loss with Focaler-DIoU is developed to address the imbalance between positive and negative samples across multiple classes. Experimental results demonstrate that MTF-NET achieves a 5.1% improvement in mAP50 on the VisDrone2019 dataset, surpassing current state-of-the-art methods, and further yields enhancements of 4.2% and 13.4% on the UA-DETRAC-G2 and HazyDet datasets, respectively. These findings effectively validate the robustness and generalization capabilities of our network, providing a potent solution for object detection in complex mixed traffic flow scenarios.<\/jats:p>","DOI":"10.1371\/journal.pone.0344151","type":"journal-article","created":{"date-parts":[[2026,3,16]],"date-time":"2026-03-16T17:44:39Z","timestamp":1773683079000},"page":"e0344151","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["MTF-NET: A mixed traffic flow multi-target detection network based on full-field perception and adaptive optimization"],"prefix":"10.1371","volume":"21","author":[{"given":"Shihao","family":"Li","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0864-8915","authenticated-orcid":true,"given":"Qiao","family":"Meng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xin","family":"Liu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhijie","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Siyuan","family":"Kong","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bingyu","family":"Li","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"340","published-online":{"date-parts":[[2026,3,16]]},"reference":[{"issue":"1","key":"pone.0344151.ref001","first-page":"1","article-title":"Review of studies on mixed traffic flow: perspective of developing economies","volume":"2","author":"A Verma","year":"2016","journal-title":"Transp in Dev Econ"},{"issue":"4","key":"pone.0344151.ref002","doi-asserted-by":"crossref","first-page":"3340","DOI":"10.1109\/COMST.2019.2924143","article-title":"Design challenges of multi-UAV systems in cyber-physical applications: a comprehensive survey and future directions","volume":"21","author":"R Shakeri","year":"2019","journal-title":"IEEE Commun Surv Tutorials"},{"issue":"3","key":"pone.0344151.ref003","doi-asserted-by":"crossref","first-page":"257","DOI":"10.1109\/JPROC.2023.3238524","article-title":"Object detection in 20 years: a survey","volume":"111","author":"Z Zou","year":"2023","journal-title":"Proc IEEE"},{"key":"pone.0344151.ref004","author":"J Redmon","year":"2018"},{"key":"pone.0344151.ref005","author":"A Bochkovskiy","year":"2020"},{"key":"pone.0344151.ref006","author":"R Khanam","year":"2024"},{"key":"pone.0344151.ref007","author":"MAR Alif","year":"2024"},{"issue":"7","key":"pone.0344151.ref008","doi-asserted-by":"crossref","first-page":"753","DOI":"10.3390\/jmse9070753","article-title":"Sea surface object detection algorithm based on YOLO v4 fused with Reverse Depthwise Separable Convolution (RDSC) for USV","volume":"9","author":"T Liu","year":"2021","journal-title":"JMSE"},{"key":"pone.0344151.ref009","doi-asserted-by":"crossref","first-page":"110063","DOI":"10.1109\/ACCESS.2021.3102600","article-title":"Adaptive fusion of multi-scale YOLO for pedestrian detection","volume":"9","author":"W-Y Hsu","year":"2021","journal-title":"IEEE Access"},{"issue":"7","key":"pone.0344151.ref010","doi-asserted-by":"crossref","first-page":"1759","DOI":"10.1049\/ipr2.13061","article-title":"An algorithm for detecting dense small objects in aerial photography based on coordinate position attention module","volume":"18","author":"H Wu","year":"2024","journal-title":"IET Image Processing"},{"issue":"8","key":"pone.0344151.ref011","doi-asserted-by":"crossref","first-page":"526","DOI":"10.3390\/drones7080526","article-title":"Drone-YOLO: An efficient neural network method for target detection in drone images","volume":"7","author":"Z Zhang","year":"2023","journal-title":"Drones"},{"key":"pone.0344151.ref012","doi-asserted-by":"crossref","first-page":"126857","DOI":"10.1109\/ACCESS.2023.3329713","article-title":"YOLO-UAV: object detection method of unmanned aerial vehicle imagery based on efficient multi-scale feature fusion","volume":"11","author":"C Ma","year":"2023","journal-title":"IEEE Access"},{"key":"pone.0344151.ref013","article-title":"Attention is all you need","volume":"30","author":"A Vaswani","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"pone.0344151.ref014","unstructured":"Chen M, Radford A, Child R, Wu J, Jun H, Luan D, et al. Generative pretraining from pixels. In: International conference on machine learning. 2020. p. 1691\u2013703."},{"key":"pone.0344151.ref015","author":"A Dosovitskiy","year":"2020"},{"key":"pone.0344151.ref016","doi-asserted-by":"crossref","unstructured":"Liu Z, Lin Y, Cao Y, Hu H, Wei Y, Zhang Z, et al. Swin transformer: hierarchical vision transformer using shifted windows. In: 2021 IEEE\/CVF International Conference on Computer Vision (ICCV). 2021. p. 9992\u201310002. doi: 10.1109\/iccv48922.2021.00986","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"pone.0344151.ref017","doi-asserted-by":"crossref","unstructured":"Yuan L, Chen Y, Wang T, Yu W, Shi Y, Jiang Z, et al. Tokens-to-Token ViT: training vision transformers from scratch on ImageNet. In: 2021 IEEE\/CVF International Conference on Computer Vision (ICCV). 2021. p. 538\u201347. doi: 10.1109\/iccv48922.2021.00060","DOI":"10.1109\/ICCV48922.2021.00060"},{"key":"pone.0344151.ref018","first-page":"24261","article-title":"Mlp-mixer: an all-mlp architecture for vision","volume":"34","author":"IO Tolstikhin","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"pone.0344151.ref019","author":"J Lee-Thorp","year":"2021"},{"issue":"4","key":"pone.0344151.ref020","doi-asserted-by":"crossref","first-page":"5314","DOI":"10.1109\/TPAMI.2022.3206148","article-title":"ResMLP: feedforward networks for image classification with data-efficient training","volume":"45","author":"H Touvron","year":"2023","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"pone.0344151.ref021","doi-asserted-by":"crossref","unstructured":"Yu W, Luo M, Zhou P, Si C, Zhou Y, Wang X, et al. MetaFormer is actually what you need for vision. In: 2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2022. p. 10809\u201319. doi: 10.1109\/cvpr52688.2022.01055","DOI":"10.1109\/CVPR52688.2022.01055"},{"key":"pone.0344151.ref022","author":"R Wightman","year":"2021"},{"issue":"13","key":"pone.0344151.ref023","doi-asserted-by":"crossref","first-page":"1389","DOI":"10.1049\/icp.2024.3085","article-title":"YOLO-Rep: lightweight UAV visual tracking algorithm","volume":"2024","author":"W Yi","year":"2025","journal-title":"IET Conf Proc"},{"key":"pone.0344151.ref024","doi-asserted-by":"crossref","first-page":"121209","DOI":"10.1016\/j.eswa.2023.121209","article-title":"YOLO-FA: type-1 fuzzy attention based YOLO detector for vehicle detection","volume":"237","author":"L Kang","year":"2024","journal-title":"Expert Systems with Applications"},{"issue":"26","key":"pone.0344151.ref025","doi-asserted-by":"crossref","first-page":"16023","DOI":"10.1007\/s00521-024-09917-2","article-title":"Position information encoding FPN for small object detection in aerial images","volume":"36","author":"D Feng","year":"2024","journal-title":"Neural Comput & Applic"},{"issue":"2","key":"pone.0344151.ref026","doi-asserted-by":"crossref","DOI":"10.32604\/cmc.2024.052138","article-title":"MCBAN: a small object detection multi-convolutional block attention network","volume":"81","author":"H Bhanbhro","year":"2024","journal-title":"Computers, Materials & Continua"},{"key":"pone.0344151.ref027","first-page":"290","volume-title":"Dq-detr: Detr with dynamic query for tiny object detection. In: European Conference on Computer Vision","author":"YX Huang","year":"2024"},{"key":"pone.0344151.ref028","doi-asserted-by":"crossref","first-page":"124848","DOI":"10.1016\/j.eswa.2024.124848","article-title":"EL-YOLO: An efficient and lightweight low-altitude aerial objects detector for onboard applications","volume":"256","author":"C Xue","year":"2024","journal-title":"Expert Systems with Applications"},{"key":"pone.0344151.ref029","doi-asserted-by":"crossref","unstructured":"Chen J, Kao S, He H, Zhuo W, Wen S, Lee CH. Run, don\u2019t walk: chasing higher FLOPS for faster neural networks. In: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition. 2023. p. 12021\u201331.","DOI":"10.1109\/CVPR52729.2023.01157"},{"key":"pone.0344151.ref030","doi-asserted-by":"crossref","unstructured":"Sandler M, Howard A, Zhu M, Zhmoginov A, Chen L-C. MobileNetV2: inverted residuals and linear bottlenecks. In: 2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition. 2018. p. 4510\u201320. doi: 10.1109\/cvpr.2018.00474","DOI":"10.1109\/CVPR.2018.00474"},{"key":"pone.0344151.ref031","doi-asserted-by":"crossref","unstructured":"Chen H, Chen K, Ding G, Han J, Lin Z, Liu L, et al. YOLOv10: real-time end-to-end object detection. In: Advances in Neural Information Processing Systems, 37, 2024. p. 107984\u20138011. doi: 10.52202\/079017-3429","DOI":"10.52202\/079017-3429"},{"key":"pone.0344151.ref032","unstructured":"Tishby N, Pereira FC, Bialek W. The information bottleneck method. In: arXiv preprint, 2000. https:\/\/arxiv.org\/abs\/physics\/0004057"},{"key":"pone.0344151.ref033","doi-asserted-by":"crossref","unstructured":"Lin T-Y, Dollar P, Girshick R, He K, Hariharan B, Belongie S. Feature pyramid networks for object detection. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. p. 936\u201344. doi: 10.1109\/cvpr.2017.106","DOI":"10.1109\/CVPR.2017.106"},{"issue":"2","key":"pone.0344151.ref034","doi-asserted-by":"crossref","first-page":"1426","DOI":"10.1609\/aaai.v38i2.27907","article-title":"Omni-Kernel network for image restoration","volume":"38","author":"Y Cui","year":"2024","journal-title":"AAAI"},{"key":"pone.0344151.ref035","first-page":"443","volume-title":"No more strided convolutions or pooling: a new CNN building block for low-resolution images and small objects. In: Joint European conference on machine learning and knowledge discovery in databases","author":"R Sunkara","year":"2022"},{"key":"pone.0344151.ref036","doi-asserted-by":"crossref","first-page":"110714","DOI":"10.1016\/j.patcog.2024.110714","article-title":"YOLO-FaceV2: a scale and occlusion aware face detector","volume":"155","author":"Z Yu","year":"2024","journal-title":"Pattern Recognition"},{"key":"pone.0344151.ref037","doi-asserted-by":"crossref","unstructured":"Cai Z, Ravichandran A, Maji S, Fowlkes C, Tu Z, Soatto S. Exponential moving average normalization for self-supervised and semi-supervised learning. In: 2021 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021. p. 194\u2013203. doi: 10.1109\/cvpr46437.2021.00026","DOI":"10.1109\/CVPR46437.2021.00026"},{"key":"pone.0344151.ref038","author":"H Zhang","year":"2024"},{"issue":"07","key":"pone.0344151.ref039","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.0344151.ref040","unstructured":"Du D, Zhu P, Wen L, Bian X, Lin H, Hu Q. VisDrone-DET2019: The vision meets drone object detection in image challenge results. In: Proceedings of the IEEE\/CVF international conference on computer vision workshops. 2019."},{"key":"pone.0344151.ref041","doi-asserted-by":"crossref","first-page":"102907","DOI":"10.1016\/j.cviu.2020.102907","article-title":"UA-DETRAC: A new benchmark and protocol for multi-object detection and tracking","volume":"193","author":"L Wen","year":"2020","journal-title":"Computer Vision and Image Understanding"},{"key":"pone.0344151.ref042","author":"C Feng","year":"2024"},{"key":"pone.0344151.ref043","doi-asserted-by":"crossref","unstructured":"Zhao Y, Lv W, Xu S, Wei J, Wang G, Dang Q, et al. DETRs beat YOLOs on real-time object detection. In: 2024 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2024. p. 16965\u201374. doi: 10.1109\/cvpr52733.2024.01605","DOI":"10.1109\/CVPR52733.2024.01605"},{"key":"pone.0344151.ref044","unstructured":"Peng Y, Li H, Wu P, Zhang Y, Sun X, Wu F. D-FINE: Redefine regression task in DETRs as fine-grained distribution refinement. 2024. doi: arXiv:241013842"},{"key":"pone.0344151.ref045","doi-asserted-by":"crossref","unstructured":"Huang S, Lu Z, Cun X, Yu Y, Zhou X, Shen X. DEIM: DETR with improved matching for fast convergence. In: 2025 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025. p. 15162\u201371. doi: 10.1109\/cvpr52734.2025.01412","DOI":"10.1109\/CVPR52734.2025.01412"},{"issue":"8","key":"pone.0344151.ref046","doi-asserted-by":"crossref","first-page":"8673","DOI":"10.1609\/aaai.v39i8.32937","article-title":"FBRT-YOLO: faster and better for real-time aerial image detection","volume":"39","author":"Y Xiao","year":"2025","journal-title":"AAAI"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0344151","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,16]],"date-time":"2026-03-16T17:44:48Z","timestamp":1773683088000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0344151"}},"subtitle":[],"editor":[{"given":"Musa","family":"Aydin","sequence":"first","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2026,3,16]]},"references-count":46,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3,16]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0344151","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,16]]}}}