{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,12]],"date-time":"2026-10-12T04:27:53Z","timestamp":1791779273884,"version":"4.3.4"},"reference-count":39,"publisher":"Public Library of Science (PLoS)","issue":"5","license":[{"start":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T00:00:00Z","timestamp":1746662400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Yangjiang Nuclear Power Domestic Intelligent Security System Construction Project"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Steel surface defect detection is an important application of object detection in industry. Achieving object detection in industry while balancing detection accuracy and real-time performance is a challenge. Therefore, this paper proposes an improved FasterNet-YOLO model based on the one-stage detector. Introduce the FasterNet network to reconstruct the YOLOv5 backbone network. Achievement of model lightweighting and significant improvement in detection speed, but with a slight reduction in accuracy. The YOLOv5 neck network\u2019s ordinary convolution is improved by depthwise separable convolution. Continuing to improve detection speed while further reducing redundant parameters in the neck network. To improve model accuracy, the Swin-Transformer is integrated into the C3 module in the neck network. Solve the problem of cluttered backgrounds in defect photographs and easy confusion between defect types. Meanwhile, BiFPN is used for feature fusion. By retaining more informative features, the detector\u2019s ability to adapt to targets at different scales is improved. The results indicated that when comparing FasterNet-YOLO with the original model, the parameters were reduced by 49.4%, GFLOPs were reduced by 57.0%, mAP increased by 6.2%, and FPS increased by 54.1%. The improved model not only increases the detection accuracy, but also significantly improves the speed of hot-rolled strip surface defect detection to meet the requirements of real-time detection.<\/jats:p>","DOI":"10.1371\/journal.pone.0323248","type":"journal-article","created":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T16:05:52Z","timestamp":1746720352000},"page":"e0323248","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":17,"title":["FasterNet-YOLO for real-time detection of steel surface defects algorithm"],"prefix":"10.1371","volume":"20","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-7359-9160","authenticated-orcid":true,"given":"Shiwei","family":"Yu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zelin","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liang","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaoqiang","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jikui","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2025,5,8]]},"reference":[{"key":"pone.0323248.ref001","first-page":"2517","article-title":"Transfer learning for automated optical inspection","author":"S Kim","year":"2017"},{"issue":"11","key":"pone.0323248.ref002","doi-asserted-by":"crossref","first-page":"8389","DOI":"10.1007\/s00521-022-08112-5","article-title":"Automated steel surface defect detection and classification using a new deep learning-based approach","volume":"35","author":"K Demir","year":"2022","journal-title":"Neural Comput & Applic"},{"key":"pone.0323248.ref003","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TIM.2020.3040485","article-title":"RetinaNet With Difference Channel Attention and Adaptively Spatial Feature Fusion for Steel Surface Defect Detection","volume":"70","author":"X Cheng","year":"2021","journal-title":"IEEE Trans Instrum Meas"},{"issue":"1","key":"pone.0323248.ref004","doi-asserted-by":"crossref","first-page":"885","DOI":"10.1007\/s40747-023-01180-7","article-title":"A deep learning model for steel surface defect detection","volume":"10","author":"Z Li","year":"2023","journal-title":"Complex Intell Syst"},{"issue":"5","key":"pone.0323248.ref005","doi-asserted-by":"crossref","first-page":"1674","DOI":"10.3390\/s24051674","article-title":"An Infrared Image Defect Detection Method for Steel Based on Regularized YOLO","volume":"24","author":"Y Zou","year":"2024","journal-title":"Sensors (Basel)"},{"key":"pone.0323248.ref006","article-title":"Outer surface defect detection of steel pipes with 3D vision based on multi-line structured lights","author":"L Jin","year":"2024","journal-title":"Measurement Science and Technology"},{"issue":"2","key":"pone.0323248.ref007","doi-asserted-by":"crossref","first-page":"303","DOI":"10.1049\/ipr2.12647","article-title":"Review of surface defect detection of steel products based on machine vision","volume":"17","author":"B Tang","year":"2022","journal-title":"IET Image Processing"},{"issue":"3","key":"pone.0323248.ref008","doi-asserted-by":"crossref","first-page":"626","DOI":"10.1109\/TIM.2019.2963555","article-title":"Automated Visual Defect Detection for Flat Steel Surface: A Survey","volume":"69","author":"Q Luo","year":"2020","journal-title":"IEEE Trans Instrum Meas"},{"issue":"8","key":"pone.0323248.ref009","doi-asserted-by":"crossref","first-page":"086004","DOI":"10.1088\/1361-6501\/ad4410","article-title":"FCS-YOLO: an efficient algorithm for detecting steel surface defects","volume":"35","author":"C Zhao","year":"2024","journal-title":"Meas Sci Technol"},{"issue":"6","key":"pone.0323248.ref010","doi-asserted-by":"crossref","first-page":"871","DOI":"10.3390\/jmse12060871","article-title":"FSN-YOLO: Nearshore Vessel Detection via Fusing Receptive-Field Attention and Lightweight Network","volume":"12","author":"N Du","year":"2024","journal-title":"JMSE"},{"key":"pone.0323248.ref011","doi-asserted-by":"crossref","first-page":"109118","DOI":"10.1016\/j.compag.2024.109118","article-title":"Faster-YOLO-AP: A lightweight apple detection algorithm based on improved YOLOv8 with a new efficient PDWConv in orchard","volume":"223","author":"Z Liu","year":"2024","journal-title":"Computers and Electronics in Agriculture"},{"key":"pone.0323248.ref012","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TIM.2023.3241994","article-title":"An ultrasmall bolt defect detection method for transmission line inspection","volume":"72","author":"P Luo","year":"2023","journal-title":"IEEE Transactions on Instrumentation and Measurement"},{"key":"pone.0323248.ref013","doi-asserted-by":"crossref","first-page":"107924","DOI":"10.1016\/j.optlaseng.2023.107924","article-title":"Defect detection of MicroLED with low distinction based on deep learning","volume":"173","author":"M Chen","year":"2024","journal-title":"Optics and Lasers in Engineering"},{"issue":"2","key":"pone.0323248.ref014","doi-asserted-by":"crossref","first-page":"2628","DOI":"10.1364\/OE.480816","article-title":"A visual defect detection for optics lens based on the YOLOv5 -C3CA-SPPF network model","volume":"31","author":"H Tang","year":"2023","journal-title":"Opt Express"},{"key":"pone.0323248.ref015","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TIM.2023.3276026","article-title":"Surface Defect Detection of Aeroengine Blades Based on Cross-Layer Semantic Guidance","volume":"72","author":"K Song","year":"2023","journal-title":"IEEE Trans Instrum Meas"},{"key":"pone.0323248.ref016","first-page":"329","article-title":"Analyzing the performance of multilayer neural networks for object recognition","author":"P Agrawal","year":"2014"},{"key":"pone.0323248.ref017","first-page":"1440","article-title":"Fast r-cnn","author":"R Girshick","year":"2015"},{"key":"pone.0323248.ref018","first-page":"779","article-title":"You only look once: Unified, real-time object detection","author":"J Redmon","year":"2016"},{"key":"pone.0323248.ref019","article-title":"Yolov4: Optimal speed and accuracy of object detection.","author":"A Bochkovskiy","year":"2020"},{"key":"pone.0323248.ref020","article-title":"YOLOv6: A single-stage object detection framework for industrial applications.","author":"C Li","year":"2022"},{"key":"pone.0323248.ref021","first-page":"7464","article-title":"YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors","author":"CY Wang","year":"2023"},{"key":"pone.0323248.ref022","first-page":"21","article-title":"Ssd: Single shot multibox detector","author":"W Liu","year":"2016"},{"key":"pone.0323248.ref023","doi-asserted-by":"crossref","first-page":"107466","DOI":"10.1016\/j.triboint.2022.107466","article-title":"Intelligent prediction of wear location and mechanism using image identification based on improved Faster R-CNN model","volume":"169","author":"M Wang","year":"2022","journal-title":"Tribology International"},{"key":"pone.0323248.ref024","first-page":"237","article-title":"SSDD-Net: A Lightweight and Efficient Deep Learning Model for Steel Surface Defect Detection","author":"Z Li","year":"2023"},{"key":"pone.0323248.ref025","doi-asserted-by":"crossref","first-page":"112776","DOI":"10.1016\/j.measurement.2023.112776","article-title":"RDD-YOLO: A modified YOLO for detection of steel surface defects","volume":"214","author":"C Zhao","year":"2023","journal-title":"Measurement"},{"key":"pone.0323248.ref026","doi-asserted-by":"crossref","first-page":"122256","DOI":"10.1016\/j.eswa.2023.122256","article-title":"A YOLO-NL object detector for real-time detection","volume":"238","author":"Y Zhou","year":"2024","journal-title":"Expert Systems with Applications"},{"key":"pone.0323248.ref027","article-title":"Mobilenets: Efficient convolutional neural networks for mobile vision applications.","author":"A G Howard","year":"2017"},{"key":"pone.0323248.ref028","first-page":"116","article-title":"Shufflenet v2: Practical guidelines for efficient cnn architecture design","author":"N Ma","year":"2018"},{"key":"pone.0323248.ref029","first-page":"1580","article-title":"Ghostnet: More features from cheap operations","author":"K Han","year":"2020"},{"key":"pone.0323248.ref030","article-title":"Contextnet: Exploring context and detail for semantic segmentation in real-time.","author":"R P K Poudel","year":"2018"},{"key":"pone.0323248.ref031","first-page":"392","article-title":"A hierarchical graph network for 3d object detection on point clouds","author":"J Chen","year":"2020"},{"key":"pone.0323248.ref032","first-page":"12021","article-title":"Run, Don\u2019t walk: Chasing higher FLOPS for faster neural networks","author":"J Chen","year":"2023"},{"issue":"04","key":"pone.0323248.ref033","first-page":"5117","article-title":"Pconv: The missing but desirable sparsity in dnn weight pruning for real-time execution on mobile devices","volume":"34","author":"X Ma","year":"2020"},{"key":"pone.0323248.ref034","first-page":"1251","article-title":"Xception: Deep learning with depthwise separable convolutions","author":"F Chollet","year":"2017"},{"key":"pone.0323248.ref035","first-page":"10012","article-title":"Swin transformer: Hierarchical vision transformer using shifted windows","author":"Z Liu","year":"2021"},{"key":"pone.0323248.ref036","first-page":"10781","article-title":"Efficientdet: Scalable and efficient object detection","author":"M Tan","year":"2020"},{"key":"pone.0323248.ref037","first-page":"2117","article-title":"Feature pyramid networks for object detection","author":"T Y Lin","year":"2017"},{"key":"pone.0323248.ref038","first-page":"8759","article-title":"Path aggregation network for instance segmentation","author":"S Liu","year":"2018"},{"key":"pone.0323248.ref039","doi-asserted-by":"crossref","first-page":"294","DOI":"10.1016\/j.optlaseng.2019.06.020","article-title":"Semi-supervised defect classification of steel surface based on multi-training and generative adversarial network","volume":"122","author":"Y He","year":"2019","journal-title":"Optics and Lasers in Engineering"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0323248","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T16:05:58Z","timestamp":1746720358000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0323248"}},"subtitle":[],"editor":[{"given":"Mohammed","family":"Abdelsamea,","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2025,5,8]]},"references-count":39,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2025,5,8]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0323248","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5,8]]}}}