{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,22]],"date-time":"2026-06-22T18:51:42Z","timestamp":1782154302190,"version":"3.54.5"},"reference-count":36,"publisher":"Springer Science and Business Media LLC","issue":"10","license":[{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T00:00:00Z","timestamp":1778630400000},"content-version":"vor","delay-in-days":12,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100007642","name":"Universit\u00e0 degli Studi di Roma Tor Vergata","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100007642","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Neural Comput &amp; Applic"],"published-print":{"date-parts":[[2026,5]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Automatic traffic sign detection is a key capability for intelligent vehicles and a fundamental component in the smart management of road infrastructure and traffic maintenance systems. Accurate, large-scale monitoring in real-world conditions requires models that balance high performance with computational efficiency. This study presents a lightweight, deployable version of YOLOv8-n, trained using a multi-source data augmentation strategy to support applications such as surveillance networks and intelligent transportation systems. To improve spatial, contextual, and illumination diversity in the training data, three independent augmentation methods were applied: (1) structured cropping of regions of interest (ROIs), (2) synthetic image generation using varied background scenes, and (3) the Copy-Paste technique incorporating real traffic sign instances. These augmented datasets were incrementally combined with real-world images from the DFG dataset to construct a diverse and realistic training corpus. All models were trained using an input resolution of 640\u2009\u00d7\u2009640 pixels. Extensive evaluations on five distinct validation datasets showed that the proposed model achieved over 90% mAP@50 (mean Average Precision at an Intersection over Union threshold of 50%) accuracy in most scenarios, including out-of-distribution conditions. With its low computational cost and efficient inference, the model is well-suited for real-time applications such as traffic sign condition monitoring, infrastructure maintenance planning, and road safety improvement. These findings support the development of cost-effective, accurate, and scalable solutions for intelligent traffic management systems.<\/jats:p>","DOI":"10.1007\/s00521-026-12087-y","type":"journal-article","created":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T06:16:31Z","timestamp":1778652991000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A data-centric multi-source augmentation framework for deployable traffic sign recognition using YOLOv8-n"],"prefix":"10.1007","volume":"38","author":[{"given":"Hamed","family":"Mortazavinasab","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Simone","family":"Celli","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aldo","family":"Di Carlo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,5,13]]},"reference":[{"key":"12087_CR1","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s40537-019-0197-0","volume":"6","author":"C Shorten","year":"2019","unstructured":"Shorten C, Khoshgoftaar TM (2019) A survey on image data augmentation for deep learning. J Big Data 6:1\u201348","journal-title":"J Big Data"},{"key":"12087_CR2","doi-asserted-by":"publisher","first-page":"683","DOI":"10.1016\/j.ijtst.2022.06.002","volume":"12","author":"RK Megalingam","year":"2023","unstructured":"Megalingam RK, Thanigundala K, Musani SR et al (2023) Indian traffic sign detection and recognition using deep learning. Int J Transp Sci Technol 12:683\u2013699","journal-title":"Int J Transp Sci Technol"},{"key":"12087_CR3","doi-asserted-by":"publisher","first-page":"323","DOI":"10.1016\/j.neunet.2012.02.016","volume":"32","author":"J Stallkamp","year":"2012","unstructured":"Stallkamp J, Schlipsing M, Salmen J, Igel C (2012) Man vs. computer: Benchmarking machine learning algorithms for traffic sign recognition. Neural Netw 32:323\u2013332","journal-title":"Neural Netw"},{"key":"12087_CR4","unstructured":"Bochkovskiy A, Wang C-Y, Liao H-YM (2020) Yolov4: Optimal speed and accuracy of object detection. arXiv preprint arXiv:200410934"},{"key":"12087_CR5","doi-asserted-by":"publisher","first-page":"7853","DOI":"10.1007\/s00521-022-08077-5","volume":"35","author":"J Wang","year":"2023","unstructured":"Wang J, Chen Y, Dong Z, Gao M (2023) Improved YOLOv5 network for real-time multi-scale traffic sign detection. Neural Comput Appl 35:7853\u20137865","journal-title":"Neural Comput Appl"},{"key":"12087_CR6","doi-asserted-by":"publisher","first-page":"297","DOI":"10.3390\/math12020297","volume":"12","author":"M Flores-Calero","year":"2024","unstructured":"Flores-Calero M, Astudillo CA, Guevara D et al (2024) Traffic sign detection and recognition using YOLO object detection algorithm: A systematic review. Mathematics 12:297","journal-title":"Mathematics"},{"key":"12087_CR7","unstructured":"Ultralytics (2023) YOLOv8 Documentation. In: https:\/\/docs.ultralytics.com"},{"key":"12087_CR8","doi-asserted-by":"publisher","first-page":"100057","DOI":"10.1016\/j.array.2021.100057","volume":"10","author":"A Gupta","year":"2021","unstructured":"Gupta A, Anpalagan A, Guan L, Khwaja AS (2021) Deep learning for object detection and scene perception in self-driving cars: Survey, challenges, and open issues. Array 10:100057","journal-title":"Array"},{"key":"12087_CR9","doi-asserted-by":"crossref","unstructured":"Merolla D, Latorre V, Salis A, Boanelli G (2025) Improving road safety with AI: automated detection of signs and surface damage. Computers 14(3):91.","DOI":"10.3390\/computers14030091"},{"key":"12087_CR10","doi-asserted-by":"publisher","first-page":"125","DOI":"10.3390\/info11020125","volume":"11","author":"A Buslaev","year":"2020","unstructured":"Buslaev A, Iglovikov VI, Khvedchenya E et al (2020) Albumentations: fast and flexible image augmentations. Information 11:125","journal-title":"Information"},{"key":"12087_CR11","doi-asserted-by":"crossref","unstructured":"Yun S, Han D, Oh SJ et al (2019) Cutmix: Regularization strategy to train strong classifiers with localizable features. In: Proceedings of the IEEE\/CVF international conference on computer vision. pp 6023\u20136032","DOI":"10.1109\/ICCV.2019.00612"},{"key":"12087_CR12","doi-asserted-by":"crossref","unstructured":"Ghiasi G, Cui Y, Srinivas A et al (2021) Simple copy-paste is a strong data augmentation method for instance segmentation. In: Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition. pp 2918\u20132928","DOI":"10.1109\/CVPR46437.2021.00294"},{"key":"12087_CR13","first-page":"736","volume":"45","author":"SV Igorevich","year":"2021","unstructured":"Igorevich SV (2021) Road images augmentation with synthetic traffic signs using neural networks. \u041a\u043e\u043c\u043f\u044c\u044e\u0442\u0435\u0440\u043d\u0430\u044f \u043e\u043f\u0442\u0438\u043a\u0430 45:736\u2013748","journal-title":"\u041a\u043e\u043c\u043f\u044c\u044e\u0442\u0435\u0440\u043d\u0430\u044f \u043e\u043f\u0442\u0438\u043a\u0430"},{"key":"12087_CR14","doi-asserted-by":"crossref","unstructured":"Feng C, Zhong Y, Jie Z et al (2024) Instagen: Enhancing object detection by training on synthetic dataset. In: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition. pp 14121\u201314130","DOI":"10.1109\/CVPR52733.2024.01339"},{"key":"12087_CR15","unstructured":"German Traffic Sign Recognition Benchmark (GTSRB) In: https:\/\/benchmark.ini.rub.de"},{"key":"12087_CR16","unstructured":"LISA Traffic Sign Dataset In: http:\/\/cvrr.ucsd.edu"},{"key":"12087_CR17","doi-asserted-by":"crossref","unstructured":"Zhu Z, Liang D, Zhang S et al (2016) Traffic-sign detection and classification in the wild. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp 2110\u20132118","DOI":"10.1109\/CVPR.2016.232"},{"key":"12087_CR18","doi-asserted-by":"publisher","first-page":"1427","DOI":"10.1109\/TITS.2019.2913588","volume":"21","author":"D Tabernik","year":"2019","unstructured":"Tabernik D, Sko\u010daj D (2019) Deep learning for large-scale traffic-sign detection and recognition. IEEE Trans Intell Transp Syst 21:1427\u20131440","journal-title":"IEEE Trans Intell Transp Syst"},{"key":"12087_CR19","doi-asserted-by":"publisher","first-page":"7558","DOI":"10.1038\/s41598-025-88184-0","volume":"15","author":"H Zhang","year":"2025","unstructured":"Zhang H, Liang M, Wang Y (2025) YOLO-BS: a traffic sign detection algorithm based on YOLOv8. Sci Rep 15:7558","journal-title":"Sci Rep"},{"key":"12087_CR20","unstructured":"Ge J (2023) Traffic Sign Recognition Dataset and Data Augmentation. arXiv preprint arXiv:230318037"},{"key":"12087_CR21","unstructured":"DFG Traffic Sign Dataset In: https:\/\/www.vicos.si\/resources\/dfg"},{"key":"12087_CR22","unstructured":"alexmihalyk23 (2022) COCO2YOLO. In: https:\/\/github.com\/alexmihalyk23\/COCO2YOLO"},{"key":"12087_CR23","doi-asserted-by":"crossref","unstructured":"Wang C-Y, Bochkovskiy A, Liao H-YM (2021) Scaled-yolov4: Scaling cross stage partial network. In: Proceedings of the IEEE\/cvf conference on computer vision and pattern recognition. pp 13029\u201313038","DOI":"10.1109\/CVPR46437.2021.01283"},{"key":"12087_CR24","doi-asserted-by":"crossref","unstructured":"Dwibedi D, Misra I, Hebert M (2017) Cut, paste and learn: Surprisingly easy synthesis for instance detection. In: Proceedings of the IEEE international conference on computer vision. pp 1301\u20131310","DOI":"10.1109\/ICCV.2017.146"},{"key":"12087_CR25","doi-asserted-by":"crossref","unstructured":"Sekizawa A, Nakajima K (2019) Context-aware Training Image Synthesis for Traffic Sign Recognition. In: VISIGRAPP (5: VISAPP). pp 466\u2013473","DOI":"10.5220\/0007387900002108"},{"key":"12087_CR26","doi-asserted-by":"crossref","unstructured":"Taylor L, Nitschke G (2018) Improving deep learning with generic data augmentation. In: 2018 IEEE symposium series on computational intelligence (SSCI). IEEE, pp 1542\u20131547","DOI":"10.1109\/SSCI.2018.8628742"},{"key":"12087_CR27","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1016\/j.neucom.2021.03.049","volume":"447","author":"Y Liu","year":"2021","unstructured":"Liu Y, Peng J, Xue J-H et al (2021) TSingNet: Scale-aware and context-rich feature learning for traffic sign detection and recognition in the wild. Neurocomputing 447:10\u201322","journal-title":"Neurocomputing"},{"key":"12087_CR28","doi-asserted-by":"publisher","first-page":"42104","DOI":"10.1109\/ACCESS.2023.3264214","volume":"11","author":"G Wang","year":"2023","unstructured":"Wang G, Zhou K, Wang L, Wang L (2023) Context-aware and attention-driven weighted fusion traffic sign detection network. IEEE Access 11:42104\u201342112","journal-title":"IEEE Access"},{"key":"12087_CR29","first-page":"3781","volume":"12","author":"L Zhang","year":"2023","unstructured":"Zhang L, Xing Z, Wang X (2023) Background Instance-Based Copy-Paste Data Augmentation for Object Detection. Electron (Basel) 12:3781","journal-title":"Electron (Basel)"},{"key":"12087_CR30","unstructured":"Shrestha S, Li W, Zhu G, Barnes N (2024) SDI-Paste: synthetic dynamic instance copy-paste for video instance segmentation. arXiv preprint arXiv:2410.13565."},{"key":"12087_CR31","unstructured":"Zhao H, Sheng D, Bao J et al (2023) X-paste: Revisiting scalable copy-paste for instance segmentation using clip and stablediffusion. In: International Conference on Machine Learning. PMLR, pp 42098\u201342109"},{"key":"12087_CR32","doi-asserted-by":"crossref","unstructured":"Liu, Y., Wang, J., Wang, W., Hu, Y., Wang, Y., & Xu, Y. (2024). CRADA: Cross domain object detection with cyclic reconstruction and decoupling adaptation. IEEE Transactions on Multimedia, 26, 6250-6261.\u200f","DOI":"10.1109\/TMM.2023.3347645"},{"key":"12087_CR33","doi-asserted-by":"publisher","unstructured":"Ruis FA, Liezenga AM, Heslinga FG, Ballan L, Eker TA, den Hollander RJM, van Leeuwen MC, Dijk J, Huizinga W (2024) Improving object detector training on synthetic data by starting with a strong baseline methodology. In: Mahler S, Wetzl J (eds) Synthetic data for artificial intelligence and machine learning: tools, techniques, and applications II. Proceedings of SPIE, vol. 13035, p 1303511. https:\/\/doi.org\/10.1117\/12.3013441","DOI":"10.1117\/12.3013441"},{"key":"12087_CR34","doi-asserted-by":"crossref","unstructured":"Staniszewski M, Kempski A, Marczyk M et al (2025) Searching for the ideal recipe for preparing synthetic data in the multi-object detection problem. Appl Sci (2076\u20133417) 15, 354-369","DOI":"10.3390\/app15010354"},{"key":"12087_CR35","unstructured":"COCO Consortium (2020) COCO Evaluation Metrics \u2013 Object Detection Metrics in COCO Dataset"},{"key":"12087_CR36","doi-asserted-by":"crossref","unstructured":"Zoph B, Cubuk ED, Ghiasi G et al (2020) Learning data augmentation strategies for object detection. In: Computer Vision\u2013ECCV 2020: 16th European Conference, Glasgow, UK, August 23\u201328, 2020, Proceedings, Part XXVII 16. Springer, pp 566\u2013583","DOI":"10.1007\/978-3-030-58583-9_34"}],"container-title":["Neural Computing and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-026-12087-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00521-026-12087-y","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-026-12087-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,22]],"date-time":"2026-06-22T18:24:57Z","timestamp":1782152697000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00521-026-12087-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5]]},"references-count":36,"journal-issue":{"issue":"10","published-print":{"date-parts":[[2026,5]]}},"alternative-id":["12087"],"URL":"https:\/\/doi.org\/10.1007\/s00521-026-12087-y","relation":{},"ISSN":["0941-0643","1433-3058"],"issn-type":[{"value":"0941-0643","type":"print"},{"value":"1433-3058","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5]]},"assertion":[{"value":"26 July 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 March 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 May 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}],"article-number":"402"}}