改进YOLOv3的金属表面缺陷检测研究

doi:10.3778/j.issn.1002-8331.2104-0324

摘要/Abstract

摘要：

针对金属表面缺陷检测中目标尺寸小和特征不清晰导致漏检的问题，提出一种改进YOLOv3的金属缺陷检测算法。在YOLOv3网络结构的基础上，将第11层浅层特征与网络深层特征融合，生成一个新的尺度为104×104特征图层，提取更多小缺陷目标特征。加入DIoU边框回归损失，为边界框提供移动方向以及更准确的位置信息，加快模型收敛。利用K-Means++聚类分析数据集上的先验框尺寸信息，筛选出最优的Anchor Box，使定位更加精准，降低网络损失。将改进后的算法与其他检测算法在NEU-DET数据集上进行检测性能对比。实验分析表明改进后的YOLOv3平均检测速率为31.6?frame/s；平均检测精度为67.64%，比YOLOv3提高了7.49个百分点，相较于Faster R-CNN等算法也有较大的检测精度优势。结论表明，改进后的YOLOv3可以使小缺陷目标的位置信息和精度更加准确。

关键词: 目标检测, 金属表面缺陷, YOLOv3, K-Means++, 距离交并比（DIoU）

Abstract:

Aiming at the problems of missing detection caused by small target size and unclear features in metal surface defect detection, an improved metal defect detection algorithm based on YOLOv3 is proposed. Firstly, the shallow features of the 11th layer are fused with the deep features of the network to generate a new scale of 104×104 feature layer on the basis of the network structure of YOLOv3. Its purpose is to extract more features of small defects. Secondly, it uses DIoU as the bounding box regression loss function to provide the moving direction and more accurate position information for the bounding box to accelerate the convergence of model. Finally, K-Means++ is used to analyze the size information of the Anchor Box on the dataset. The optimal Anchor Box is selected to make the positioning more accurate and reduce the network loss. The performance of the improved algorithm is compared with other detection algorithms on NEU-DET dataset. Experimental analysis shows that the average detection rate of the improved YOLOv3 is 31.6 frame/s. Its average detection accuracy is 67.64%, which is 7.49 percentage points higher than that of YOLOv3. Compared with Faster R-CNN and other algorithms, it also has a greater detection accuracy advantage. The results show that the improved YOLOv3 can make the location information and accuracy of small defect target more accurate.

Key words: target detection, metal surface defects, YOLOv3, K-Means++, Distance Intersection over Union（DIoU）

程婧怡，段先华，朱伟. 改进YOLOv3的金属表面缺陷检测研究[J]. 计算机工程与应用, 2021, 57(19): 252-258.

CHENG Jingyi, DUAN Xianhua, ZHU Wei. Research on Metal Surface Defect Detection by Improved YOLOv3[J]. Computer Engineering and Applications, 2021, 57(19): 252-258.

参考文献

[1] 方叶祥，甘平，陈俐.金属表面缺陷检测的改进YOLOv3算法研究[J].机械科学与技术，2020，39（9）：1390-1394.
FANG Y X，GAN P，CHEN L.Improved YOLOv3 algorithm for detection of metal surface defect[J].Mechanical Science and Technology for Aerospace Engineering，2020，39（9）：1390-1394.
[2] 徐镪，朱洪锦，范洪辉，等.改进的YOLOv3网络在钢板表面缺陷检测研究[J].计算机工程与应用，2020，56（16）：265-272.
XU X，ZHU H J，FAN H H，et al.Study on detection of steel plate surface defects by improved YOLOv3 network[J].Computer Engineering and Applications，2020，56（16）：265-272.
[3] TAO X，ZHANG D，MA W，et al.Automatic metallic surface defect detection and recognition with convolutional neural networks[J].Applied Sciences-Basel，2018，8（9）.
[4] HE K，GKIOXARI G，DOLLAR P，et al.Mask R-CNN[J].IEEE Transactions on Pattern Analysis & Machine Inte-lligence，2020，42：386-397.
[5] 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，2015，39（6）：1137-1149.
[6] REDMON J，DIVVALA S，GIRSHICK R，et al.You Only Look Once：Unified，real-time object detection[C]//Proceedings of IEEE International Conference on Computer Vision & Pattern Recognition，2016：779-788.
[7] LIU W，ANGUELOV D，ERHAN D，et al.SSD：Single Shot MultiBox Detector[C]//Proceedings of European Conference on Computer Vision，2016：21-37.
[8] 汤勃，孔建益，伍世虔.机器视觉表面缺陷检测综述[J].中国图象图形学报，2017，22（12）：1640-1663.
TANG B，KONG J Y，WU S Q.Review of surface defect detection based on machine vision[J].Journal of Image and Graphics，2017，22（12）：1640-1663.
[9] 黄凤荣，李杨，郭兰申，等.基于Faster R-CNN的零件表面缺陷检测算法[J].计算机辅助设计与图形学学报，2020，32（6）：30-40.
HUANG F R，LI Y，GUO L S，et al.Method for detecting surface defects of engine parts based on Faster R-CNN[J].Journal of Computer Aided Design and Graphics，2020，32（6）：30-40.
[10] 张磊，郎贤礼，王乐.基于图像融合与YOLOv3的铝型材表面缺陷检测[J].计算机与现代化，2020，303（11）：12-19.
ZHANG L，LANG X L，WANG L.Surface defect detection of aluminum profile based on image fusion and YOLOv3[J].Computer and Modernization，2020，303（11）：12-19.
[11] 袁野，谭晓阳.复杂环境下的冰箱金属表面缺陷检测[J].计算机应用，2021，41（1）：270-274.
YUAN Y，TAN X Y.Defect detection of refrigerators metal surface in complex environment[J].Computer Applications，2021，41（1）：270-274.
[12] 李维刚，叶欣，赵云涛，等.基于改进YOLOv3算法的带钢表面缺陷检测[J].电子学报，2020，48（7）：1284-1292.
LI W G，YE X，ZHAO Y T，et al.Strip steel surface defect detection based on improved YOLOV3 algorithm[J].Acta Electronica Sinica，2020，48（7）：1284-1292.
[13] REDMON J，FARHADI A.YOLOv3：An incremental improvement[J].arXiv：1804.02767，2018.
[14] ZHENG Z，WANG P，LIU W，et al.Distance-IoU loss：Faster and better learning for bounding box regression[C]//Proceedings of the AAAI Conference on Artifical Inte-lligence，2019.
[15] TRESHANSKY A，MCGRAW R M.Overview of clustering algorithms[C]//Proceedings of the International Society for Optical Engineering，2001.
[16] REDMON J，FARHADI A.YOLO9000：Better，faster，stronger[C]//IEEE Conference on Computer Vision & Pattern Recognition，2017：6517-6525.
[17] LONG J，SHELLHAMER E，DARRELL T.Fully convolutional networks for semantic segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2015，39（4）：640-651.
[18] HE K，ZHANG X，REN S，et al.Deep residual learning for image recognition[C]//IEEE Conference on Computer Vision & Pattern Recognition，2016：770-778.
[19] 鞠默然，罗海波，王仲博，等.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）：245-252.
JU M R，LUO H B，WANG Z B，et al.Improved YOLOV3 algorithm and its application in small target detection[J].Acta Optica Sinica，2019，39（7）：245-252.
[20] KRIZHEVSKY A，SUTSKEVER I，HINTON G.ImageNet classification with deep convolutional neural networks[C]//Proceedings of NIPS，2012：1097-1105.
[21] VOUROS A，VASILAKI E.A semi-supervised sparse K-Means algorithm[J].Pattern Recognition Letters，2020，142：65-71.
[22] 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，2020，69（14）：1493-1504.