当前位置: 首页 > 文章 > 基于卷积神经网络的机械零件表面缺陷分类 农业装备与车辆工程 2019,57 (11) 19-23
Position: Home > Articles > Classification of Surface Defects of Mechanical Parts Based on Convolution Neural Network Agricultural Equipment & Vehicle Engineering 2019,57 (11) 19-23

基于卷积神经网络的机械零件表面缺陷分类

作  者:
周明浩;朱家明
关键词:
卷积神经网络;机械零件;表面缺陷;分类
摘  要:
针对人工检测缺陷时间慢和错检率高的缺点,提出将卷积神经网络应用于机械零件表面缺陷检测,该方法首先对图像进行预处理操作,然后利用图像块数据集用卷积神经网络进行训练,研究网络参数,确定网络模型,最后测试.实验结果表明,该方法在机械零件缺陷检测上面具有较为理想的准确率.
译  名:
Classification of Surface Defects of Mechanical Parts Based on Convolution Neural Network
作  者:
Zhou Minghao;Zhu Jiaming;College of Mechanical Engineering, University of Shanghai for Science and Technology;
关键词:
convolution neural network;;machine part;;surface imperfection;;classify
摘  要:
Aiming at the shortcomings of slow detection and high error detection rate, the convolution neural network is applied to surface defect detection of mechanical parts. Firstly, the image preprocessing operation is carried out. Then the image block data set is trained with convolution neural network to study the network parameters, determine the network model, and finally test. The experimental results show that the method has an ideal accuracy rate on defect detection of mechanical parts.

相似文章

计量
文章访问数: 7
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊