长江大学计算机科学学院;
针对目前钢材表面缺陷检测方法存在检测精度不高，易出现误检、漏检等问题，提出一种改进YOLOv5的钢材表面缺陷检测算法。在主干网络中引入坐标注意力(Coordinate Attention, CA)机制模块，提升模型关注钢材表面缺陷的能力，使用GhostBottleneck结构与主干网络中的部分卷积模块和C3模块进行替换，构建轻量化模型；在Neck层采用双向特征金字塔网络(Bidirectional Feature Pyramid Network, BiFPN)结构来提升检测效果；增加一个目标检测层来解决数据集中部分缺陷占比较大的问题。实验结果表明，改进的YOLOv5s-GCBD(GhostBottleneck-CA-BiFPN-Anchor)算法在NEU-DET数据集上的平均精度均值(mean Average Precision, mAP)达到80.2%,较原YOLOv5s算法提高了3.5%。相比传统的钢材表面缺陷检测方法，提出的算法实现了更精准的钢材表面缺陷检测。
| 1,222 | 8 | 47 |
| 下载次数 | 被引频次 | 阅读次数 |
[1] 王红星，陈玉权，张欣，等.基于离线高斯模型的输电线路无人机巡检缺陷智能识别方法研究[J].电测与仪表，2022,59(3):92-99.
[2] LI Y D,ZHANG S S,WANG W Q.A Lightweight Faster R-CNN for Ship Detection in SAR Images[J].IEEE Geoscience and Remote Sensing Letters,2020,19:1-5.
[3] WANG Y W,RAO Y B,HUANG C,et al.Using the Improved Mask R-CNN and Softer-NMS for Target Segmentation of Remote Sensing Image[C]//2021 4th International Conference on Pattern Recognition and Artificial Intelligence (PRAI).Yibin:IEEE,2021:91-95.
[4] CAI L,DONG F,CHEN K,et al.An FPGA Based Heterogeneous Accelerator for Single Shot MultiBox Detector (SSD)[C]//2020 IEEE 15th International Conference on Solid-State & Integrated Circuit Technology (ICSICT).Kunming:IEEE,2020:1-3.
[5] ARAKI R,ONISHI T,HIRAKAWAT,et al.MT-DSSD:Deconvolutional Single Shot Detector Using Multi Task Learning for Object Detection,Segmentation,and Grasping Detection[C]//2020 IEEE International Conference on Robotics and Automation (ICRA).Paris:IEEE,2020:10487-10493.
[6] REDMON J,DIVVALA S,GIRSHICK R,et al.You Only Look Once:Unified,Real-time Object Detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE,2016:779-788.
[7] REDMON J,FARHADI A.YOLO9000:Better,Faster,Stronger[C]//2017 IEEE Conference On Computer Vision and Pattern Recognition.Honolulu:IEEE,2017:6517-6525.
[8] REDMON J,FARHADI A.YOLOv3:An Incremental Improvement[EB/OL].(2018-04-08)[2023-05-06].https://arxiv.org/abs/1804.02767.
[9] 范加利，田少兵，黄葵，等.基于Faster R-CNN的航母舰面多尺度目标检测算法[J].系统工程与电子技术，2022,44(1):40-46.
[10] 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.
[11] 邹旺，吉畅.一种改进YOLOv4-tiny的带钢表面缺陷实时检测方法[J/OL].机械科学与技术：1-7[2023-03-28].https://doi.org/10.13433/j.cnki.1003-8728.20230034.
[12] 李鑫，汪诚，李彬，等.改进YOLOv5的钢材表面缺陷检测算法[J].空军工程大学学报(自然科学版),2022,23(2):26-33.
[13] 叶欣.基于深度学习的热轧带钢表面缺陷检测算法研究[D].武汉：武汉科技大学，2021.
[14] LIN T Y,DOLLáR P,GIRSHICK R,et al.Feature Pyramid Networks for Object Detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition.Honolulu:IEEE,2017:2117-2125.
[15] LIU S,QI L,QIN H F,et al.Path Aggregation Network for Instance Segmentation[C]//2018 IEEE Conference on Computer Vision and Pattern Recognition.Salt Lake City:IEEE,2018:8759-8768.
[16] HAN K,WANG Y H,TIAN Q,et al.Ghostnet:More Features from Cheap Operations[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Seattle:IEEE,2020:1580-1589.
[17] HOU Q B,ZHOU D Q,FENG J S.Coordinate Attention for Efficient Mobile Network Design[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Nashville:IEEE,2021:13708-13717.
[18] 石玉诚.基于深度学习的显著性目标检测研究与应用[D].贵阳：贵州大学，2022..
[19] TAN M X,PANG R M,LE Q V.EfficientDet:Scalable and Efficient Object Detection[C]//2020 IEEE/CVF Confe-rence on Computer Vision and Pattern Recognition.Seattle:IEEE,2020:10778-10787.
基本信息:
DOI：
中图分类号:TP391.41;TG115
引用信息:
[1]徐明升,祝俊辉,干家欣等.基于YOLOv5的钢材表面缺陷检测算法[J].无线电工程,2024,54(02):351-359.
基金信息:
湖北省科技示范项目(2019ZYYD016)~~