摘要
对输入图像缺陷进行准确定位和分类,通过一个紧凑的卷积神经网络(CNN)将分割结果的缺陷区域划分为特定的类别。使用工业数据集可以成功地检测出各种条件下的金属缺陷。
To accurately locate and classify the defects in the collected input images,we apple a compact convolutional neural network(CNN)to put the defect areas of the segmentation results into the specific categories.Using industrial data sets,the metal defects under various conditions can be detected.
作者
张淇雅
逄焕利
ZHANG Qiya;PANG Huanli(School of Computer Science and Engineering, Changchun University of Technology, Changchun 130012, China)
出处
《长春工业大学学报》
CAS
2021年第2期160-167,共8页
Journal of Changchun University of Technology
基金
吉林省科技厅重点科技研发基金资助项目(20180201129GX)。
关键词
缺陷自动检测
卷积神经网络
目标检测
分类
automatic defect detection
convolutional neural network
target detection
classification
