基于卷积神经网络的短切毡缺陷分类  

Classification of Chopped Strand Mat Defects Based on Convolutional Neural Network

作　　者：卓东 景军锋[1] 张缓缓[1] 苏泽斌[1] Zhuo Dong;Jing Junfeng;Zhang Huanhuan;Su Zebin(School of Electrotiic Information, Xi'an Polytechnic University, Xi'an , Shaanxi 710048, China)

机构地区：[1]西安工程大学电子信息学院

出　　处：《激光与光电子学进展》2019年第10期136-143,共8页Laser & Optoelectronics Progress

基　　金：国家自然科学基金(61301276);陕西省重点研发计划(2017GY-003);陕西省高校科协青年人才托举计划项目(20180115);陕西省教育厅科研计划项目(18JK0338)

摘　　要：基于卷积神经网络,提出了短切毡缺陷分类的方法。通过旋转、平移和翻转对数据集进行扩充,解决了小数据样本在深度卷积神经网络中的过拟合问题;利用迁移学习的思想加速网络收敛,提高了网络的泛化能力;对比了不同网络结构并选择较好的网络进行数据集验证。结果表明,所提方法能够实现短切毡缺陷的有效分类,准确率为93%。In this study, a classification method of chopped strand mat defects based on convolutional neural network is proposed. In the proposed method, the rotation, translation, and inversion are employed to expand the dataset for solving the overfitting problem caused by the small data samples in the deep convolutional neural networks. Transfer learning is employed to improve the convergence speed and generalization ability of the network. Further, the different network structures are compared, and the most optimal network structure is used to verify the database. The experimental results demonstrate that the proposed method can effectively classify the chopped strand mat defects with an accuracy rate of 93%.

关 键 词：图像处理 卷积神经网络 缺陷分类 泛化能力 短切毡 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]