摘要
针对热轧带钢表面缺陷检测中存在的检测速度慢、检测精度低等问题,提出了一种改进的YOLOv3算法模型.使用加权K-means聚类算法来优化确定先验框参数,提高先验框(priors anchor)与特征图层(feature map)的匹配度;同时,调整YOLOv3算法的网络结构,融合浅层特征与深层特征,形成新的大尺度检测图层,提高网络对带钢表面缺陷的检测精度.实验结果表明,改进后的YOLOv3算法在NEU-DET数据集上平均精度均值达到了80%,较原有的YOLOv3算法提高了11%;同时检测速度保持在50fps,优于目前其它深度学习带钢表面缺陷检测算法.
To solve the problem of slow speed and low accuracy in the surface defect detection of hot rolled strips,an improved YOLOv3 algorithm is proposed.Firstly,the weighting K-means clustering algorithm is put forward to optimize priors anchor’s parameters,which can improve the match between priors anchor and feature map.Secondly,the improved network structure of the YOLOv3 algorithm is proposed to improve the detection accuracy,whose shallow features and deep features are combined to form the new large-scale inspection layer.The experiments are carried out on the NEU-DET dataset,the results show that the average accuracy of the improved YOLOv3 algorithm is 80%,which is 11%higher than that of the original algorithm;the detection speed is 50fps,which is faster than other strip surface defect detection algorithms based on deep learning.
作者
李维刚
叶欣
赵云涛
王文波
LI Wei-gang;YE Xin;ZHAO Yun-tao;WANG Wen-bo(Engineering Research Center for Metallurgical Automation and Technology of Ministry of Education,Wuhan University of Science and Technology,Wuhan,Hubei 430081,China;National-Provincial Joint Engineering Research Center of High Temperature Materials and Lining Technology,Wuhan University of Science and Technology,Wuhan,Hubei 430081,China;College of Science,Wuhan University of Science and Technology Wuhan,Hubei 430081,China)
出处
《电子学报》
EI
CAS
CSCD
北大核心
2020年第7期1284-1292,共9页
Acta Electronica Sinica
基金
国家自然科学基金(No.51774219)。
