改进YOLOv4算法的航空发动机部件表面缺陷检测  

Surface Defect Detection of Aeroengine Components Based on Improved YOLOv4 Algorithm

作　　者：李彬 汪诚[1] 吴静[1] 刘吉超 童立甲 郭振平 Li Bin;Wang Cheng;Wu Jing;Liu Jichao;Tong Lijia;Guo Zhenping(Fundamentals Department,Air Force Engineering University,Xi'an,Shaanxi 710038,China)

机构地区：[1]空军工程大学基础部,陕西西安710038

出　　处：《激光与光电子学进展》2021年第14期406-415,共10页Laser & Optoelectronics Progress

基　　金：国家自然科学基金(52004295)。

摘　　要：针对传统方式检测航空发动机部件表面缺陷存在检测精度低、检测速度慢的问题,提出了一种改进YOLOv4算法的航空发动机部件表面缺陷检测方法。构建航空发动机部件表面缺陷数据集,使用K-means算法对缺陷样本进行聚类,获得不同大小的先验框参数;利用改进的参数调整算法对先验框尺寸进行缩放,加大先验框尺寸差异,提高先验框与特征层之间的匹配度;在主干特征提取网络输出的不同特征层后和空间金字塔池化结构后增加卷积层,提高网络对缺陷特征的提取能力。实验结果表明,改进后的YOLOv4算法在测试集上的平均精度均值(mAP)高达82.67%,比原始的YOLOv4算法提高了4.55个百分点,单张图片的平均检测时间为0.1240 s,与原始算法检测时间基本持平,检测性能也优于Faster R-CNN和YOLOv3。Aiming at solving the problems of low accuracy and slow speed in the surface defects detection of aeroengine components using traditional methods,an improved YOLOv4 algorithm is proposed herein.First,an aeroengine component surface defects dataset was developed and the K-means clustering algorithm was suggested to cluster the defect samples for obtaining the priori anchor’s parameters of different sizes.Second,the improved parameter-adjustment algorithm was used to scale the priori anchor’s sizes and increase the difference in sizes to improve the matching rate between priori anchors and feature maps.Finally,convolution layers were added after the different feature layers of the backbone feature extraction network output and spatial pyramid pooling structure to improve the ability of network to extract defect features.Experimental results show that the mean average precision(mAP)value of the improved YOLOv4 algorithm in the test dataset is as high as 82.67%,which is 4.55 percent point greater than that of the original YOLOv4 algorithm.The average detection time of a single image is 0.1240 s,which is basically the same as that of the original algorithm.Moreover,the detection performance is better than Faster R-CNN and YOLOv3.

关 键 词：机器视觉 YOLOv4 航空发动机 表面缺陷检测 深度学习 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]