基于改进VGG13的冲压件表面缺陷识别方法研究  

Research on Surface Defect Identification Method of Stamped Parts Based on Improved VGG13

作　　者：刘荣光 朱传军[1] 成佳闻 王林琳[1] LIU Rongguang;ZHU Chuanjun;CHENG Jiawen;WANG Linlin(School of Mechanical Engineering,Hubei University of Technology,Wuhan Hubei 430068,China;State Key Laboratory of Digital Manufacturing Equipment and Technology,Huazhong University of Science and Technology,Wuhan Hubei 430074,China)

机构地区：[1]湖北工业大学机械工程学院,湖北武汉430068 [2]华中科技大学数字制造装备与技术国家重点实验室,湖北武汉430074

出　　处：《机床与液压》2024年第2期199-203,共5页Machine Tool & Hydraulics

基　　金：国家自然科学基金国际(地区)合作与交流项目(51861165202);广东省重点领域研发计划项目(2019B090921001);湖北工业大学博士科研启动基金项目(BSQD2019010)。

摘　　要：针对现有冲压件制品缺陷检测方法准确率低的问题,分析深度学习的原理与方法,以VGG13网络为基准模型,通过在特征提取层之后增加CBAM模块进行改进,提出5种基于VGG13与CBAM注意力机制模块相结合的网络模型(VGG13-CBAM),将改进后的新模型与改进前原VGG13模型分别在武汉某制造车间采集的冲压件缺陷数据集上进行实验研究。将数据集以6∶2∶2划分为训练集、验证集、测试集,并使用数据增强进一步扩充训练集,增加模型泛化性能,对比数据增强前后效果的提升。实验结果表明:在改进后的VGG13-CBAM03网络与VGG13-CBAM04网络上效果明显提升,测试集正确率由79.65%分别提高到了81.55%和81.40%,在使用数据增强对训练集进行扩充后,测试集正确率分别达到84.25%和84.15%,有效提升了冲压件缺陷检测准确率。In view of the low accuracy of the existing defect detection methods for stamped parts,the principle and method of deep learning were analyzed,and the VGG13 network was used as the reference model to improve by adding CBAM modules after the feature extraction layer,and five kinds of network model based on VGG13 and CBAM were proposed.The network model(VGG13-CBAM)combined with the attention mechanism module was experimentally studied on the stamped parts defect dataset collected by a manufac⁃turing workshop in Wuhan with the improved new models and the original VGG13 model.The dataset was divided into training set,vali⁃dation set,and test set according to 6∶2∶2,and data enhancement was used to further expand the training set,the generalization per⁃formance of the model was increased,and the improvement effects before and after data enhancement were compared.The experimental results show that the effect is significantly improved on the improved VGG13-CBAM03 network and VGG13-CBAM04 network,and the test sets accuracy rate is increased from 79.65%to 81.55%and 81.40%,respectively;after expanding the training set by using data enhancement,the accuracy of the test sets reach 84.25%and 84.15%,respectively,which effectively improves the accuracy of stamped parts defect detection.

关 键 词：冲压件缺陷识别 VGG13 数据增强 CBAM模块 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP391[自动化与计算机技术—控制科学与工程]