[1]马保亮,李建达,李 仟,等.基于改进YOLOv3的螺纹钢表面缺陷检测算法[J].工业仪表与自动化装置,2025,(03):87-92.[doi:10.19950/j.cnki.CN61-1121/TH.2025.03.016]
　MA Baoliang,LI Jianda,LI Qian,et al.Rebar defect detection algorithm based on improved YOLOv3[J].Industrial Instrumentation & Automation,2025,(03):87-92.[doi:10.19950/j.cnki.CN61-1121/TH.2025.03.016]
点击复制

基于改进YOLOv3的螺纹钢表面缺陷检测算法()

《工业仪表与自动化装置》[ISSN:1000-0682/CN:61-1121/TH]

卷:
期数:
2025年03期
页码:
87-92
栏目:
出版日期:
2025-06-15

文章信息/Info

Title:
Rebar defect detection algorithm based on improved YOLOv3
文章编号:
1000-0682（2025）03-0087-06
作者:
马保亮; 李建达; 李 仟; 徐紫琪; 肖 剑
中交二公局东萌工程有限公司，陕西 西安 710064
Author(s):
MA Baoliang ;  LI Jianda;  LI Qian ;  XU Ziqi;  XIAO Jian
(CCCC Second Highway Engineering Bureau，Dongmeng Engineering，Shaanxi Xi’an 710064, China)
关键词:
缺陷检测; YOLOv3; 螺纹钢; 深度学习; k-means++
Keywords:
defect detection;  YOLOv3;  rebar;  deep learning;  k-means++
分类号:
TP391
DOI:
10.19950/j.cnki.CN61-1121/TH.2025.03.016
文献标志码:
A
摘要:
针对传统的螺纹钢表面缺陷检测方法效率低、检测精度差、漏检误检率高等问题，提出了一种基于YOLOv3改进的螺纹钢表面缺陷检测算法。首先在主干网络中引入了强化模型特征提取能力的密集连接方式，有效避免了信息在多层传递过程中的丢失和模糊，确保了信息在深度网络结构中依然能够保持清晰和有效地表达，提高了小目标缺陷的检测精确度；利用k-means++算法对螺纹钢的缺陷数据集进行重聚类分析，确保预设锚框能更精确地匹配缺陷的多尺度特性；在损失函数中采用SIoU算法，在改善预测框与真实框不重合时无法梯度回传问题的同时加快收敛速度；最后在预测阶段采用softer-NMS算法替换NMS算法，能够有效优化边界框的位置回归，从而提高对重叠目标的辨识率。实验结果表明，在自建的螺纹钢数据集中，改进后的网络使模型的平均精度均值mAP在保持较高FPS的情况下提高了7.7%，其指标能够满足实时检测的需求，针对螺纹钢常见的锈迹、裂痕、擦伤等7种表面缺陷，其平均检测精度达到87.4%。
Abstract:
In order to address the challenges of low efficiency, poor detection accuracy, and high rates of missed and false detections in traditional methods for detecting surface defects in rebar, we propose an improved defect detection algorithm based on YOLOv3.First, we introduce a dense connection method in the backbone network to enhance the model’s feature extraction capabilities. This effectively prevents the loss and blurriness of information during multi-layer transmission, ensuring that information remains clear and effectively expressed in the deep network structure. This improvement increases the detection accuracy of small defects.Next, we employ the k-means++ algorithm for re-clustering analysis of the rebar defect dataset, ensuring that the predefined anchor boxes can more accurately match the multi-scale characteristics of the defects.We also adopt the SIoU algorithm in the loss function. This not only addresses the gradient backpropagation issue when the predicted boxes do not overlap with the ground truth boxes but also accelerates the convergence speed.Finally, during the prediction phase, we replace the NMS algorithm with the softer-NMS algorithm. This effectively optimizes the bounding box regression, improving the identification rate of overlapping targets.Experimental results show that, on our self-built rebar dataset, the improved network increases the model’s mAP by 7.7% while maintaining a high FPS, meeting the requirements for real-time detection. The average detection accuracy for seven common surface defects, including rust, cracks, and scratches, reaches 88.7%.

参考文献/References:

[1] 陈永峰.低合金螺纹钢腐蚀机理与耐蚀性调控研究现状[J].中国冶金,2024,34(06):36-43.

[2] 建辉,孙显彬,刘伦明,等.基于漏磁检测的矿用钢丝绳励磁装置[J].工矿自动化,2023,49(07):114-119.
[3] 王瑾,李勇,苏冰洁,等.基于双传感器差分机制的铁磁性金属板件亚表面腐蚀缺陷脉冲涡流可视化检测[J].中国机械工程, 2024:1-11.
[4] 李烨斌,张书增,赵宝,等.服役钢轨轨底缺陷的超声无损检测方案及装置设计[J/OL].中国铁路,2024,(09):195-201.
[5] 孙鸽,张运楚,赵月,等.基于机器视觉的螺纹钢表面缺陷检测方法[J].计算机系统应用,2020,29(04):32-40.
[6] Redmon J , Divvala S , Girshick R ,et al.You Only Look Once: Unified, Real-Time Object Detection[C]//Computer Vision & Pattern Recognition.IEEE, 2016.
[7] 李和平,陈中举,许浩然,等.改进YOLO的钢材表面缺陷检测算法[J].现代电子技术,2024,47(13):7-14.
[8] Redmon J, Farhadi A. Yolov3: An incremental improvement[J]. arxiv preprint arxiv : 1804.02767, 2018.
[9] Lin T Y, Dollar P, Girshick R, et al. Feature Pyramid Networks for Object Detection [C]. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
[10] Redmon J, Farhadi A. YOLO9000: Better, Faster, Stronger [C]. IEEE, 2017: 6517-6525.
[11] H?m?l?inen J, K?rkk?inen T, Rossi T. Improving scalable K-means++[J]. Algorithms, 2020, 14(1): 6.
[12] Huang G, Liu Z, Laurens V, et al. Densely Connected Convolutional Networks [C]. Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 4700-4708.
[13] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770-778.
[14] He Y, Zhang X, Savvides M, et al. Softer-nms: Rethinking bounding box regression for accurate object detection [J]. arXiv preprint arXiv: 1809. 08545, 2018, 2(3).
[15] Zhou D, Fang J, Song X, et al. IoU loss for 2d/3d object detection [C]. 2019 International Conference on 3D Vision (3DV). IEEE, 2019: 85-94.
[16] Rezatofighi H, Tsoi N, Gwak J, et al. Generalized Intersection Over Union: A Metric and a Loss for Bounding Box Regression [C]. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019.
[17] Gevorgyan Z. SIoU loss: More powerful learning for bounding box regression[J]. arXiv preprint arXiv: 2205.12740, 2022.

相似文献/References:

[1]吴全玉,周 署,俞 洋,等.Hough算法在印刷电路板焊孔缺陷检测中的应用[J].工业仪表与自动化装置,2015,(06):56.
　WU Quanyu,ZHOU Shu,YU Yang,et al.Research Hough algorithm of detection technology for PCB holes defaults[J].Industrial Instrumentation & Automation,2015,(03):56.
[2]杨林超,张新锋*,刘 康.管道机器人采集图像缺陷检测方法研究[J].工业仪表与自动化装置,2021,(06):48.[doi:10.19950/j.cnki.cn61-1121/th.2021.06.026]
　YANG Linchao,ZHANG Xinfeng *,LIU Kang.Research on defect detection method of pipeline robot image acquisition[J].Industrial Instrumentation & Automation,2021,(03):48.[doi:10.19950/j.cnki.cn61-1121/th.2021.06.026]
[3]王明吉a,刘 博a,陈秋梦a,等.基于yolov3的车牌定位识别系统[J].工业仪表与自动化装置,2022,(01):97.[doi:10.19950/j.cnki.cn61-1121/th.2022.01.020]
　WANG Mingjia,LIU Boa,CHEN Qiumenga,et al.License plate location and recognition system based on yolov3[J].Industrial Instrumentation & Automation,2022,(03):97.[doi:10.19950/j.cnki.cn61-1121/th.2022.01.020]
[4]王建冲,高军伟,张炳星,等.基于机器视觉的SOP芯片引脚缺陷检测系统[J].工业仪表与自动化装置,2022,(05):32.[doi:10.19950/j.cnki.cn61-1121/th.2022.05.006]
　WANG Jianchong,GAO Junwei,ZHANG Bingxing,et al.Design of SOP chip pin defect detection system based on machine vision[J].Industrial Instrumentation & Automation,2022,(03):32.[doi:10.19950/j.cnki.cn61-1121/th.2022.05.006]
[5]单文轩,徐永祥,白福忠,等.基于改进YOLOv3的工业指针式仪表检测[J].工业仪表与自动化装置,2023,(04):64.[doi:10.19950/j.cnki.cn61-1121/th.2023.04.012]
　SHAN Wenxuan,XU Yongxiang,BAI Fuzhong,et al.Industrial pointer meter detection based on improved YOLOv3[J].Industrial Instrumentation & Automation,2023,(03):64.[doi:10.19950/j.cnki.cn61-1121/th.2023.04.012]
[6]何军红,温观发,黎长鑫.优化Faster R-CNN算法的小样本缺陷检测研究[J].工业仪表与自动化装置,2024,(05):94.[doi:DOI:10.19950/j.cnki.cn61-1121/th.2024.05.018]
　HE Junhong,WEN Guanfa,LI Changxin.Research on small sample defect detection based on optimized Faster R-CNN algorithm[J].Industrial Instrumentation & Automation,2024,(03):94.[doi:DOI:10.19950/j.cnki.cn61-1121/th.2024.05.018]

备注/Memo

备注/Memo:
收稿日期：2024-10-25第一作者：马保亮（1979—），男，陕西渭南人，高级工程师，主要从事智能建造及检验检测技术方面工作。E-mail：275399251@qq.com
更新日期/Last Update: 1900-01-01