锻压技术

基于YOLOv4算法的冲压件缺陷检测

英文标题：Defect detection of stamping parts based on YOLOv4 algorithm

作者：孙永鹏1 2 钟佩思1 2 刘梅2 曹爱霞3 李梁1 2

单位：1. 山东科技大学先进制造技术研究中心 2. 山东科技大学机械电子工程学院 3. 青岛黄海学院智能制造学院

分类号：TP391.41；TG38

出版年,卷(期):页码：2022,47(1):222-228

摘要：

针对冲压件缺陷检测目前存在的人工检测强度大、效率低等问题，提出了一种基于改进YOLOv4(You Only Look Once)模型的快速检测算法（YOLOv4-Mobile）。该方法使用改进的MobileNetV3网络代替YOLOv4结构中的CSPDarknet53网络，改进的MobileNetV3网络结合了深度可分离卷积、具有线性瓶颈的倒残差结构以及SE结构（轻量级注意力结构）。利用车间采集的冲压件图像，建立缺陷数据集并进行数据增强，使用K均值（K-means）聚类算法得到一组对应冲压件缺陷数据集的先验框参数，提高了先验框与特征图层的匹配度。实验结果表明：基于改进YOLOv4模型的快速检测算法的平均精度达到89%，高于SSD算法；同时，单张检测时间达到0.15 s，优于原有的YOLOv4算法。

For the problems of high manual detection intensity and low efficiency in defect detection of stamping parts at present, a fast detection algorithm (YOLOv4-Mobile) based on the improved YOLOv4 (You Only Look Once) model was proposed, which used the improved MobileNetV3 network to replace the CSPDarknet53 network in YOLOv4 structure, and the improved MobileNetV3 network combined a depthwise separable convolution, an inverted residual structure with a linear bottleneck and SE (Squeeze and Excitation) structure. Then, the image of stamping parts collected in the workshop was used to establish the defect data set and enhance the data set, and a set of prior frame parameters corresponding to the defect data set of stamping parts was obtained by K-means clustering algorithm to improve the matching degree of prior frame and feature layer. The test results show that based on the improved YOLOv4 model, the mAP(mean Average Precision) of the fast detection algorithm reaches 89%, which is higher than that of SSD algorithm. Meanwhile, the detection speed reaches 0.15 s per sheet, which is better than the original YOLOv4 algorithm.

基金项目：

山东省重点研发计划资助项目(2019GGX104102)；山东省自然科学基金资助项目(ZR2017MEE066)

作者简介：孙永鹏(1995-)，男，硕士研究生 E-mail：sunyongpenglvlife@163.com 通信作者：钟佩思(1966-)，男，博士，教授 E-mail：pszhong@163.com

参考文献：

[1] 邓凡,刘彦强,樊建中,等.基于数字图像相关技术的泡沫铝复合结构的弯曲行为研究[J].稀有金属,2021,45(3):297-305.

Deng F, Liu Y Q, Fan J Z, et al. Observation of bending behavior of aluminum foam composite structure based on digital image correlation technology[J]. Chinese Journal of Rare Metals, 2021,45(3):297-305.

[2] 陶显,侯伟,徐德.基于深度学习的表面缺陷检测方法综述[J/OL].自动化学报:1-19[2020-10-11]. http://kns.cnki.net/kcms/detail/11.2109.TP.20200402.1101.002.html.

Tao X, Hou W, Xu D. A survey of surface defectdetection methods based on deep learning[J/OL]. Acta Automatica Sinica: 1-19[2020-10-11]. http://kns.cnki.net/kcms/detail/11.2109. TP.20200402.1101.002.html.

[3] 李兰,奚舒舒,张才宝,等.基于改进SSD模型的工件表面缺陷识别算法[J].计算机工程与科学,2020,42(9):1608-1615.

Li L, Xi S S, Zhang C B, et al. A surface defect recognition algorithm based on improver SSD model[J]. Computer Engineering and Science, 2020, 42(9): 1608-1615.

[4] 袁野,谭晓阳.复杂环境下的冰箱金属表面缺陷检测[J/OL].计算机应用:1-6[2020-09-22].

Yuan Y, Tan X Y. Defect detection of refrigerator metal surface in complex environment[J].Journal of Computer Applications: 1-6[2020-09-22].

[5] 李春霖,谢刚,王银,等.基于YOLOv3-TinyD算法的偏光片缺陷检测[J/OL].计算机集成制造系统:1-17[2020-09-23].

Li C L, Xie G, Wang Y, et al. Defect detection of polaroid based on YOLOv3-TinyD[J/OL]. Computer Interated Manufacturing Systems: 1-17[2020-09-23].

[6] Redmon J, Divvala S, Girshick R, et al. You only look once: Unified, real-time object detection[A]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C]. Washington D. C. : IEEE Press, 2016.

[7] Bochkovskiy A, Wang C Y, Liao H Y M. YOLOv4: Optimal speed and accuracy of object detection[EB/OL]. https://github.com/AlexeyAB/darknet, 2020.

[8] He K, Zhang X, Ren S, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence,2015，37（9）：1904-1916.

[9] Liu S, Qi L, Qin H, et al. Path aggregation network for instance segmentation[A]. IEEE/CVF Conference on Computer Vision and Pattern Recognition[C]. Washington D. C. : IEEE Press, 2018.

[10]Howard A, Sandler M, Chen B, et al. Searching for MobileNetV3[A]. 2019 IEEE/CVF International Conference on Computer Vision (ICCV)[C]. Seoul: IEEE, 2020.

[11]Elsken T, Metzen J H, Hutter F. Neural architecture search: A survey[J].Journal of Machine Learning Research,2019,20:1-21.

[12]Howard A G, Zhu M, Chen B, et al. MobileNets: Efficient convolutional neural networks for mobile vision applications[A]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C]. Hawaii: IEEE, 2017.

[13]Sandler M, Howard A, Zhu M, et al. MobileNetV2: Inverted residuals and linear bottlenecks[A]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C]. Salt Lake City: IEEE, 2018.

[14]Hu J, Shen L, Sun G, et al. Squeeze-and-excitation networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 42(8):7132-7141.

[15]Liu W, Anguelov D, Erhan D, et al. SSD: Single shot multi box detector[A]. Proc of European Conference on Computer Vision[C]. Berlin: Springer, 2016.

服务与反馈：

【文章下载】【加入收藏】