基于改进YOLOv4的电机端盖缺陷检测  

Defect Detection of Motor Cover Based on Improved YOLOv4

作　　者：万卓 叶明[1] 刘凯[1] WAN Zhuo;YE Ming;LIU Kai(College of Mechanical and Electrical Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)

机构地区：[1]南京航空航天大学机电学院,南京210016

出　　处：《计算机系统应用》2021年第3期79-87,共9页Computer Systems & Applications

基　　金：国家自然科学基金(51405229);江苏省自然科学基金(BK20151470)。

摘　　要：在基于机器视觉实现电机端盖裂纹缺陷检测过程中,针对复杂背景下目标特征不明显的问题,使用限制对比度的自适应直方图均衡化的方法加强目标特征.针对机器视觉系统中训练数据量少且训练图片背景单一导致模型泛化性低的问题,对比了Mosaic和CutMix数据增强方法,并结合多种数据增强策略,提出了系统的数据集构建方案.针对使用YOLOv4进行单类检测和小目标检测时正负样本不平衡导致检测率低的问题,提出了自适应多尺度焦点损失+CIoU损失的加权融合损失函数,并通过实验得到最优超参数.最后使用K-means算法初始化锚点框,使模型更适应线状目标的预测.结果表明,该方法对于裂纹类别的检测达到了95.8%的平均精度(Average Precision,AP),相较于改进前提升9.7%,单张检测时间48 ms,具有一定的工程应用价值.Adaptive histogram equalization with limited contrast is applied to strengthening the target feature to solve the problem of unclear targets in a complex background during crack detection of motor covers based on machine vision. A systematic dataset construction scheme is proposed by comparing Mosaic and CutMix data augmentation and combining with a variety of data enhancement techniques to address the low generalization of the model induced by the small volume of training data in the machine vision system and single background of training pictures. Besides, a weighted fusion loss function combined with adaptive multi-scale focus loss and CIoU loss is proposed to deal with the low detection rate caused by unbalanced numbers of positive and negative samples in the single class detection and small target detection of YOLOv4, and the optimal hyper parameters are obtained through experiments. Finally, the anchor box is initialized by the K-means algorithm to make the model more suitable for predicting linear targets. Results demonstrate that this method achieves an Average Precision(AP) of 95.8% for detecting crack types, which is 9.7% higher than before, and the singlesheet detection time is 48 ms, presenting the potential for engineering application.

关 键 词：机器视觉 缺陷检测 YOLOv4 数据增强 焦点损失 

分 类 号：TM31[电气工程—电机] TP391.41[自动化与计算机技术—计算机应用技术]