{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T03:03:07Z","timestamp":1787022187407,"version":"build-2736575974"},"reference-count":30,"publisher":"Public Library of Science (PLoS)","issue":"4","license":[{"start":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T00:00:00Z","timestamp":1776211200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Program of Yunnan Provincial Department of Science and Technology","award":["202301BD070001-257"],"award-info":[{"award-number":["202301BD070001-257"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["51968065"],"award-info":[{"award-number":["51968065"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>To address the problem of insufficient detection accuracy for dense targets, small targets and partially occluded objects in complex road scenarios, an improved object detection model, SSH-YOLO, is proposed. On the basis of YOLOv8n, the model optimizes and improves performance through a three-level collaborative architecture: 1) introduce the spatial and deep conversion (SPDConv) module in the backbone network to replace the traditional step downsampling with nonstep convolution, retain the fine-grained features of small targets, and solve the feature loss problem of low-resolution images; 2) embed the spatial and channel collaborative attention module (SCSA), through cross-scale feature fusion (SMSA) and channel weight progressive optimization (PCSA), focus on the key visible areas of the occluded target and suppress background interference such as roadside vegetation; and 3) add a new 160\u2009\u00d7\u2009160 resolution small object detection head, combined with the original P3\u2012P5 layer to form a four-level detection system, covering long-distance small targets\u2009&lt;\u200932\u2009\u00d7\u200932 pixels. The experimental results show that the improved model performs well on the self-built RoadScene-Complex dataset and four public datasets BDD100K: 0.729 (12.4% higher than YOLOv8n) on the RoadScene-Complex dataset mAP@0.5 (12.4% higher than YOLOv8n) and 0.868 (7.6% higher than the KITTI dataset) mAP@0.5). COCO small target subset mAP@0.5 to 0.585 (up 16.5%), and CityPersons occluded scene mAP@0.5 to 0.739 (up 22.8%). At the same time, it maintains lightweight characteristics and has an inference speed of up to 60 FPS to meet the needs of real-time on-board detection. The research results provide a balanced solution of \u201caccuracy-speed-lightweight\u201d for high-precision target detection in complex traffic scenarios, especially in small target and occlusion scenarios.<\/jats:p>","DOI":"10.1371\/journal.pone.0343924","type":"journal-article","created":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T18:04:54Z","timestamp":1776276294000},"page":"e0343924","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":1,"title":["SSH-YOLO: YOLOv8 improved model based on Object Detection in Complex Road Scenes"],"prefix":"10.1371","volume":"21","author":[{"given":"Tenglong","family":"Ma","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5704-9759","authenticated-orcid":true,"given":"Yanlin","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiaqiang","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haisheng","family":"Yu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chao","family":"He","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,4,15]]},"reference":[{"key":"pone.0343924.ref001","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1007\/s11263-019-01247-4","article-title":"Deep Learning for Generic Object Detection: A Survey","volume":"128","author":"L Liu","year":"2020","journal-title":"Int J Comput Vis"},{"issue":"7","key":"pone.0343924.ref002","doi-asserted-by":"crossref","first-page":"4316","DOI":"10.1109\/TITS.2020.3032227","article-title":"Deep Learning for Safe Autonomous Driving: Current Challenges and Future Directions","volume":"22","author":"K Muhammad","year":"2021","journal-title":"IEEE Trans Intell Transport Syst"},{"key":"pone.0343924.ref003","doi-asserted-by":"crossref","first-page":"453","DOI":"10.1016\/j.aej.2025.03.020","article-title":"A multi-objective dynamic detection model in autonomous driving based on an improved YOLOv8","volume":"122","author":"C Li","year":"2025","journal-title":"Alexandria Engineering Journal"},{"issue":"1","key":"pone.0343924.ref004","article-title":"A Novel Faster RCNN Model Based on Multi-scale Feature Fusion and Shape Priori for Dense Vehicle Detection","volume":"24","author":"Y Li","year":"2023","journal-title":"Sens Imaging"},{"key":"pone.0343924.ref005","first-page":"380","article-title":"Remote Sensing Image Detection Based on FasterRCNN. in Artificial Intelligence in China (eds. Liang, Q. et al.)","volume":"653","author":"S Liu","year":"2021"},{"issue":"1","key":"pone.0343924.ref006","doi-asserted-by":"crossref","DOI":"10.1186\/s40537-021-00434-w","article-title":"Comparative analysis of deep learning image detection algorithms","volume":"8","author":"S Srivastava","year":"2021","journal-title":"J Big Data"},{"key":"pone.0343924.ref007","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1007\/s11760-023-02726-5","article-title":"FasterNet-SSD: a small object detection method based on SSD model","volume":"18","author":"F Yang","year":"2024","journal-title":"SIViP"},{"key":"pone.0343924.ref008","doi-asserted-by":"crossref","first-page":"96","DOI":"10.1109\/DCABES52998.2021.00031","article-title":"SSD small object detection algorithm based on feature enhancement and sample selection","author":"Z Liu","year":"2021","journal-title":"2021 20th International Symposium on Distributed Computing and Applications for Business Engineering and Science (DCABES)"},{"key":"pone.0343924.ref009","first-page":"1","article-title":"Comparison of Faster-RCNN, YOLO, and SSD for real-time vehicle type recognition","author":"J Kim","year":"2020","journal-title":"2020 IEEE International Conference on Consumer Electronics - Asia (ICCE-Asia)"},{"key":"pone.0343924.ref010","doi-asserted-by":"crossref","first-page":"7020","DOI":"10.23919\/CCC50068.2020.9189180","article-title":"Covered Vehicle Detection in Autonomous Driving Based on Faster RCNN","author":"J Fan","year":"2020","journal-title":"2020 39th Chinese Control Conference (CCC), Shenyang, China"},{"key":"pone.0343924.ref011","doi-asserted-by":"crossref","first-page":"111655","DOI":"10.1016\/j.measurement.2022.111655","article-title":"Fast vehicle detection algorithm in traffic scene based on improved SSD","volume":"201","author":"Z Chen","year":"2022","journal-title":"Measurement"},{"issue":"16","key":"pone.0343924.ref012","doi-asserted-by":"crossref","first-page":"7190","DOI":"10.3390\/s23167190","article-title":"UAV-YOLOv8: A Small-Object-Detection Model Based on Improved YOLOv8 for UAV Aerial Photography Scenarios","volume":"23","author":"G Wang","year":"2023","journal-title":"Sensors (Basel)"},{"key":"pone.0343924.ref013","doi-asserted-by":"crossref","first-page":"332","DOI":"10.1007\/s42452-024-06029-3","article-title":"Improved vehicle detection: unveiling the potential of modified YOLOv5","volume":"6","author":"MdM Rana","year":"2024","journal-title":"Discov Appl Sci"},{"key":"pone.0343924.ref014","first-page":"374","article-title":"Improving Object Detection in YOLOv8n with the C2f-f Module and Multi-Scale Fusion Reconstruction","author":"H Li","year":"2024","journal-title":"2024 IEEE 6th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)"},{"key":"pone.0343924.ref015","doi-asserted-by":"crossref","first-page":"1066","DOI":"10.1016\/j.procs.2022.01.135","article-title":"A Review of Yolo Algorithm Developments","volume":"199","author":"P Jiang","year":"2022","journal-title":"Procedia Computer Science"},{"key":"pone.0343924.ref016","first-page":"1","article-title":"YOLOv8: A Novel Object Detection Algorithm with Enhanced Performance and Robustness","author":"R Varghese","year":"2024","journal-title":"2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS)"},{"key":"pone.0343924.ref017","doi-asserted-by":"crossref","first-page":"129844","DOI":"10.1016\/j.neucom.2025.129844","article-title":"Improved YOLOv7 for small object detection in airports: Task-oriented feature learning with Gaussian Wasserstein loss and attention mechanisms","volume":"634","author":"R Peng","year":"2025","journal-title":"Neurocomputing"},{"key":"pone.0343924.ref018","doi-asserted-by":"crossref","first-page":"127447","DOI":"10.1016\/j.eswa.2025.127447","article-title":"SO-YOLOv8: A novel deep learning-based approach for small object detection with YOLO beyond COCO","volume":"280","author":"Iqra","year":"2025","journal-title":"Expert Systems with Applications"},{"key":"pone.0343924.ref019","article-title":"Improved lightweight rebar detection network based on YOLOv8s algorithm","volume":"7","year":"2023","journal-title":"ACSS"},{"key":"pone.0343924.ref020","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1016\/j.isatra.2023.09.001","article-title":"ISA: Ingenious Siamese Attention for object detection algorithms towards complex scenes","volume":"143","author":"L Liu","year":"2023","journal-title":"ISA Trans"},{"key":"pone.0343924.ref021","author":"R Sunkara","year":"2022","journal-title":"No more strided convolutions or pooling: A new CNN building block for low-resolution images and small objects"},{"issue":"8","key":"pone.0343924.ref022","first-page":"8123","article-title":"A survey of deep learning-based object detection in autonomous driving","volume":"24","author":"X Zhang","year":"2023","journal-title":"IEEE Transactions on Intelligent Transportation Systems"},{"issue":"10","key":"pone.0343924.ref023","doi-asserted-by":"crossref","first-page":"105023","DOI":"10.1088\/1361-6501\/ad633d","article-title":"Enhanced feature extraction YOLO industrial small object detection algorithm based on receptive-field attention and multi-scale features","volume":"35","author":"H Tao","year":"2024","journal-title":"Meas Sci Technol"},{"key":"pone.0343924.ref024","author":"Y Si","year":"2024","journal-title":"SCSA: Exploring the Synergistic Effects Between Spatial and Channel Attention"},{"issue":"16","key":"pone.0343924.ref025","doi-asserted-by":"crossref","first-page":"2524","DOI":"10.3390\/math12162524","article-title":"Separable CenterNet Detection Network Based on MobileNetV3\u2014An Optimization Approach for Small-Object and Occlusion Issues","volume":"12","author":"Z Jiao","year":"2024","journal-title":"Mathematics"},{"key":"pone.0343924.ref026","doi-asserted-by":"crossref","first-page":"105485","DOI":"10.1016\/j.imavis.2025.105485","article-title":"Dense small target detection algorithm for UAV aerial imagery","volume":"156","author":"S Lu","year":"2025","journal-title":"Image and Vision Computing"},{"key":"pone.0343924.ref027","first-page":"740","article-title":"Microsoft COCO: Common Objects in Context","author":"T-Y Lin","year":"2014","journal-title":"Lecture Notes in Computer Science. Springer International Publishing"},{"key":"pone.0343924.ref028","doi-asserted-by":"crossref","first-page":"4457","DOI":"10.1109\/CVPR.2017.474","article-title":"CityPersons: A diverse dataset for pedestrian detection","author":"S Zhang","year":"2017","journal-title":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA"},{"key":"pone.0343924.ref029","article-title":"BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning","author":"F Yu","year":"2018"},{"key":"pone.0343924.ref030","doi-asserted-by":"crossref","first-page":"3354","DOI":"10.1109\/CVPR.2012.6248074","article-title":"Are we ready for autonomous driving? The KITTI vision benchmark suite","author":"A Geiger","year":"2012","journal-title":"2012 IEEE Conference on Computer Vision and Pattern Recognition"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0343924","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T18:05:02Z","timestamp":1776276302000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0343924"}},"subtitle":[],"editor":[{"given":"Sen","family":"Xiang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,4,15]]},"references-count":30,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,4,15]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0343924","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,15]]}}}