浏览全部资源
扫码关注微信
辽宁大学 信息学部,辽宁 沈阳 110036
沙小奇(2000-),男,辽宁大学信息学部硕士研究生,研究方向为深度学习
陈廷伟(1974-),男,博士,辽宁大学信息学部教授,研究方向为人工智能、深度学习、区块链。
收稿:2025-02-26,
纸质出版:2026-04-15
移动端阅览
沙小奇,黄建,陈廷伟.YOLO-SBS:一种改进YOLOv8s的钢材表面缺陷检测算法[J].软件导刊,2026,25(04):76-82.
SHA Xiaoqi,HUANG Jian,CHEN Tingwei.YOLO-SBS: An Improved YOLOv8s Algorithm for Steel Surface Defect Detection[J].Software Guide,2026,25(04):76-82.
沙小奇,黄建,陈廷伟.YOLO-SBS:一种改进YOLOv8s的钢材表面缺陷检测算法[J].软件导刊,2026,25(04):76-82. DOI: 10.11907/rjdk.251119.
SHA Xiaoqi,HUANG Jian,CHEN Tingwei.YOLO-SBS: An Improved YOLOv8s Algorithm for Steel Surface Defect Detection[J].Software Guide,2026,25(04):76-82. DOI: 10.11907/rjdk.251119.
针对钢材表面缺陷检测任务存在的参数量大、特征融合能力不足和小目标缺陷检测精度不佳等问题,提出一种改进YOLOv8s的钢材表面缺陷检测算法YOLO-SBS。首先,设计SCB模块改进YOLOv8中的C2f模块,提升模型特征提取能力,降低参数量;其次,引入加权特征融合WFF,依靠可学习权重动态调整特征图的贡献比例,强化模型特征融合能力;最后,引入针对小目标检测的SPDConv模块,缓解深层网络信息丢失,提升模型对缺陷检测的整体性能。在NEU-DET数据集上的实验表明,所提模型的mAP为75.0%,相较于原始的YOLOv8s提升3.7%;参数量、计算量分别降低9.9%、6.7%,实现了参数量、计算量与检测精度之间的平衡,在钢材表面缺陷检测任务中表现出色。在工业数据集GC10-DET中的泛化性验证实验表明,改进算法具有较好的泛化性。
Aiming at the problems of large parameter quantity, insufficient feature fusion ability, and poor accuracy of small target defect detection in steel surface defect detection tasks, an improved YOLO-SBS algorithm for steel surface defect detection based on YOLOv8s is proposed. Firstly, design the SCB module to improve the C2f module in YOLOv8, enhance the model's feature extraction capability, and reduce the number of parameters; Secondly, weighted feature fusion (WFF) is introduced to dynamically adjust the contribution ratio of feature maps based on learnable weights, enhancing the model's feature fusion capability; Finally, the SPDConv module for small object detection is introduced to alleviate information loss in deep networks and improve the overall performance of the model for defect detection. Experiments on the NEU-DET dataset show that the mAP of the proposed model is 75.0%, which is 3.7% higher than the original YOLOv8s; The parameter quantity and calculation quantity were reduced by 9.9% and 6.7% respectively, achieving a balance between parameter quantity, calculation quantity, and detection accuracy, and performing excellently in the task of steel surface defect detection. The generalization verification experiment on the industrial dataset GC10-DET shows that the improved algorithm has good generalization performance.
ZHANG Y , FENG F . Analysis of strip surface defect detection technology [J]. Information & Computer (Theory Edition) , 2021 , 33 ( 11 ): 19 - 22 .
张艳 , 冯锋 . 带钢表面缺陷检测技术探析 [J]. 信息与电脑(理论版) , 2021 , 33 ( 11 ): 19 - 22 .
REN S , HE K , GIRSHICK R , et al . Faster R-CNN: towards real-time object detection with region proposal networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2015 , 28 : 91 - 99 .
LIU W , ANGUELOV D , ERHAN D , et al . SSD: single shot multibox detector [C]// 14th European Conference on Computer Vision , 2016 : 21 - 37 . doi: 10.1007/978-3-319-46448-0_2 http://dx.doi.org/10.1007/978-3-319-46448-0_2
REDMON J , DIVVALA S , GIRSHICK R , et al . You only look once: unified, real-time object detection [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2016 : 779 - 788 . doi: 10.1109/cvpr.2016.91 http://dx.doi.org/10.1109/cvpr.2016.91
YANG L , ZHANG Y N , WANG T T , et al . New method for steel surface defect detection based on improved Faster R-CNN [J]. Journal of Jilin University (Information Science Edition) , 2021 , 39 ( 4 ): 409 - 415 . doi: 10.3969/j.issn.1671-5896.2021.04.008 http://dx.doi.org/10.3969/j.issn.1671-5896.2021.04.008
杨莉 , 张亚楠 , 王婷婷 , 等 . 基于改进Faster R-CNN的钢材表面缺陷检测方法 [J]. 吉林大学学报(信息科学版) , 2021 , 39 ( 4 ): 409 - 415 . doi: 10.3969/j.issn.1671-5896.2021.04.008 http://dx.doi.org/10.3969/j.issn.1671-5896.2021.04.008
YAN X , YANG Y C , TU N W . Steel surface defect detection based on improved SSD [J]. Modern Manufacturing Engineering , 2023 ( 5 ): 112 - 120 .
阎馨 , 杨月川 , 屠乃威 . 基于改进SSD的钢材表面缺陷检测 [J]. 现代制造工程 , 2023 ( 5 ): 112 - 120 .
HUANG H X , WU D . Steel defect detection based on improved YOLOv7 [J]. Computer Science , 2024 , 51 ( S1 ): 535 - 539 . doi: 10.11896/jsjkx.230800018 http://dx.doi.org/10.11896/jsjkx.230800018
黄海新 , 吴迪 . 基于改进YOLOv7的钢材缺陷检测 [J]. 计算机科学 , 2024 , 51 ( S1 ): 535 - 539 . doi: 10.11896/jsjkx.230800018 http://dx.doi.org/10.11896/jsjkx.230800018
HUANG X , WU L , LI Y , et al . Lightweight steel surface defect detection algorithm based on improved YOLOX-s [J]. Journal of Computer Applications , 2023 , 43 ( S2 ): 201 - 208 .
黄啸 , 吴龙 , 黎尧 , 等 . 基于改进YOLOX-s的轻量级型钢表面缺陷检测算法 [J]. 计算机应用 , 2023 , 43 ( S2 ): 201 - 208 .
ZHANG W K , LIU J . Steel surface defect detection based on improved YOLOv8s [J]. Journal of Beijing Information Science & Technology University (Natural Science Edition) , 2023 , 38 ( 6 ): 33 - 40 .
张文铠 , 刘佳 . 基于改进YOLOv8s的钢材表面缺陷检测 [J]. 北京信息科技大学学报(自然科学版) , 2023 , 38 ( 6 ): 33 - 40 .
JOCHER G , CHAURASIA A , QIU J . Ultralytics YOLO(Version 8.0 . 0 )[EB/OL]. https://github.com/ultralytics/ultralytics https://github.com/ultralytics/ultralytics .
TAN M X , PANG R M , LE Q V . EfficientDet: scalable and efficient object detection [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2020 : 10781 - 10790 . doi: 10.1109/cvpr42600.2020.01079 http://dx.doi.org/10.1109/cvpr42600.2020.01079
SUNKARA R , LUO T . No more strided convolutions or pooling: a new CNN building block for low-resolution images and small objects [C]// Joint European Conference on Machine Learning and Knowledge Discovery in Databases , 2022 : 443 - 459 . doi: 10.1007/978-3-031-26409-2_27 http://dx.doi.org/10.1007/978-3-031-26409-2_27
JOCHER G , CHAURASIA A , STOKEN A , et al . Ultralytics/yolov5: v7.0 - YOLOv5 sota realtime instance segmentation (v7.0) [EB/OL]. https://doi.org/10.5281/zenodo.7347926 https://doi.org/10.5281/zenodo.7347926 .
LIU S , QI L , QIN H , et al . Path aggregation network for instance segmentation [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2018 : 8759 - 8768 . doi: 10.1109/cvpr.2018.00913 http://dx.doi.org/10.1109/cvpr.2018.00913
LIN T Y , DOLLÁR P , GIRSHICK R , et al . Feature pyramid networks for object detection [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2017 : 936 - 944 . doi: 10.1109/cvpr.2017.106 http://dx.doi.org/10.1109/cvpr.2017.106
MA X , DAI X , BAI Y , et al . Rewrite the stars [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2024 : 5694 - 5703 . doi: 10.1109/cvpr52733.2024.00544 http://dx.doi.org/10.1109/cvpr52733.2024.00544
WOO S , PARK J , LEE J Y , et al . CBAM: convolutional block attention module[DB/OL] . https://arxiv.org/abs/1807.06521 https://arxiv.org/abs/1807.06521 . doi: 10.1007/978-3-030-01234-2_1 http://dx.doi.org/10.1007/978-3-030-01234-2_1
HOWARD A G , ZHU M , CHEN B , et al . MobileNets: efficient convolutional neural networks for mobile vision applications[DB/OL] . https://arxiv.org/abs/1704.04861 https://arxiv.org/abs/1704.04861 .
HE Y , SONG K , MENG Q , YAN Y , et al . An end-to-end steel surface defect detection approach via fusing multiple hierarchical features [J]. IEEE Transactions on Instrumentation and Measurement , 2020 , 69 ( 4 ): 1493 - 1504 . doi: 10.1109/tim.2019.2915404 http://dx.doi.org/10.1109/tim.2019.2915404
LI C Y , LI L , JIANG H L , et al . YOLOv6: a single-stage object detection framework for industrial applications[DB/OL] . https://arxiv.org/abs/2209.02976 https://arxiv.org/abs/2209.02976 .
GAO Z , LU C , LI H , et al . A corn seed spacing detection method based on image stitching and YOLOX [J]. Computers and Electronics in Agriculture , 2024 , 222 : 109087 . doi: 10.1016/j.compag.2024.109087 http://dx.doi.org/10.1016/j.compag.2024.109087
LYU W Y , XU S L , ZHAO Y , et al . DETRs beat YOLOs on real-time object detection[DB/OL] . https://arxiv.org/abs/2304.08069 https://arxiv.org/abs/2304.08069 . doi: 10.1109/cvpr52733.2024.01605 http://dx.doi.org/10.1109/cvpr52733.2024.01605
LYU X , DUAN F , JIANG J , et al . Deep metallic surface defect detection: the new benchmark and detection network [J]. Sensors , 2020 , 20 ( 6 ): 1562 . doi: 10.3390/s20061562 http://dx.doi.org/10.3390/s20061562
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
鄂公网安备42010602004949号