摘要
在煤的高分辨透射电镜(HRTEM)图像提取芳香层片过程中,通过对图像处理技术的应用,做出关于HRTEM图像像素的训练集和测试集,然后设计相应的卷积神经网络进行运算,探讨出使其拥有最大学习率和最高准确度的权重和偏置,最终得到可以解决煤的HRTEM图像的最佳卷积神经网络模型。
In the process of extracting aromatic slices from high resolution transmission electron microscopy(HRTEM)images of coal,through the application of image processing technology,the training set and test set of HRTEM image pixels are made,and then the corresponding convolution neural network is designed for calculation,and the weights and biases which make it have the highest learning rate and the highest accuracy are discussed.Finally,get the best convolution neural network model which can solve the HRTEM image of coal.
作者
李耀高
LI Yao-gao(College of Mining Engineering,Taiyuan University of Technology,Taiyuan 030024,China)
出处
《煤炭技术》
CAS
2019年第8期167-170,共4页
Coal Technology
