改进YOLOv5s网络在缺陷检测中的应用Application of Improved YOLOv5s Network in Defect Detection
王浩然;李廷会;曹玉军;黄飞江;李振彰;
摘要(Abstract):
针对目前金属表面缺陷检测过程中小目标缺陷检测存在着检测精度低、漏检率高等问题,本文基于经典单阶段目标检测网络的5s版本(YOLOv5s)提出了一种改进方法,并将其应用于金属鼓形滚子表面缺陷检测。由于金属表面存在大量像素占比小、尺度不规整、特征不明显的小目标缺陷,本方法通过自适应锚框计算出适合小目标尺寸的锚框值,并在原网络基础上构建了输出为8倍降采样的特征融合目标检测层,使之更匹配小目标检测任务,同时减小了模型体积,节约了计算时间。将高分辨率工业相机采集的金属鼓形滚子表面缺陷图像制作成数据集,用改进前后的YOLOv5s网络进行对比试验,实验结果表明改进后YOLOv5s网络对小目标缺陷检测的精度有明显提升,能有效检测小目标缺陷。
关键词(KeyWords): 金属表面缺陷;小目标检测;YOLOv5s;自适应锚框;多尺度特征融合
基金项目(Foundation): 广东省教育厅青年创新人才基金项目(自然科学)(编号:2019KQNCX067);广东省教育厅特色创新项目(编号:2019KTSCX127)
作者(Authors): 王浩然;李廷会;曹玉军;黄飞江;李振彰;
参考文献(References):
- [1] He Y,Song K,Meng Q,et al.An end-to-end steel surface defect detection approach via fusing multiple hierarchical features[J].IEEE Transactions on Instrumentation and Measurement,2019,69(4):1493-1504.
- [2] Girshick R.Fast R-CNN[C]//Proceedings of the IEEE international conference on computer vision.2015:1440-1448.
- [3] Ren S,He K,Girshick R,et al.Faster R-CNN:towards real-time object detection with region proposal networks[J].International Conference on Neural Information Processing Systems.MIT Press,2017,37(9):1137-1139.
- [4] He K,Gkioxari G,Dollar P,et al.Mask R-CNN[C]//Proceedings of the IEEE international conference on computer vision.2017:2961-2969.
- [5] 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.
- [6] Liu W,Anguelov D,Erhan D,et al.SSD:Single Shot Multibox Detector[C]//European conference on computer vision.Springer,Cham,2016:21-37.
- [7] 张顶,黄晞.基于SSD的手掌检测方法研究[J].网络新媒体技术,2020,9(4):28-36.
- [8] 梁鸿,王庆玮,张千等.小目标检测技术研究综述[J].计算机工程与应用,2021,57(1):17-28.
- [9] Yang G,Feng W,Jin J,et al.Face Mask Recognition System with YOLOV5 Based on Image Recognition[C]//2020 IEEE 6th International Conference on Computer and Communications (ICCC).IEEE,2020:1398-1404.
- [10] Ma W,Tian T,Xu H,et al.AABO:Adaptive Anchor Box Optimization for Object Detection via Bayesian Sub-sampling[C]//European Conference on Computer Vision.Springer,Cham,2020:560-575.
- [11] Yang Y,Song J,Huang Z,et al.Multi-feature fusion via hierarchical regression for multimedia analysis[J].IEEE Transactions on Multimedia,2012,15(3):572-581.
- [12] Hao W,Zhili S.Improved Mosaic:Algorithms for more Complex Images[J].Journal of Physics Conference Series,2020,1684:012094.
- [13] 鞠默然,罗海波,王仲博,何淼,常铮,惠斌.改进的YOLO V3算法及其在小目标检测中的应用[J].光学学报,2019,39(7):253-260.
- [14] Chen P,Li Y,Zhou H,et al.Detection of small ship objects using anchor boxes cluster and feature pyramid network model for SAR imagery[J].Journal of Marine Science and Engineering,2020,8(2):112.
- [15] Tian Z,Shen C,Chen H,et al.FCOS:A simple and strong anchor-free object detector[EB/OL].(2020-10-12).https://arxiv.org/abs/2006.09214v3.
- [16] Tian Z,Shen C,Chen H,et al.Fcos:Fully convolutional one-stage object detection[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2019:9627-9636.
- [17] 张舒然,武岩波,朱敏,房小芳.基于模拟遗传算法的水声时变多普勒频偏估计[J].网络新媒体技术,2020,9(6):45-50.
- [18] 王千,王成,冯振元,叶金凤.K-means聚类算法研究综述[J].电子设计工程,2012,20(7):21-24.
- [19] Lin T Y,Maire M,Belongie S,et al.Microsoft coco:Common objects in context[C]//European conference on computer vision.Springer,Cham,2014:740-755.
- [20] 施文灶,黄晞.基于局部不变特征点的高分辨率遥感影像配准[J].网络新媒体技术,2017,6(3):27-32+46.
- [21] 张慧,王坤峰,王飞跃.深度学习在目标视觉检测中的应用进展与展望[J].自动化学报,2017,43(8):1289-1305.
- [22] 杨杏丽.分类学习算法的性能度量指标综述[J].计算机科学,2021,48(8):209-219.
- [23] 戴立伟,黄山.交通场景目标检测指标优化研究[J].激光与光电子学进展,2020,57(14):344-352.