Indexing of electron back-scatter diffraction patterns using a convolutional neural network
Indexing of electron back-scatter diffraction patterns using a convolutional neural network
复制标题
DOI：
10.1016/j.actamat.2020.08.046
复制
发表时间：
2020-10-15
期刊：
影响因子：
9.4
通讯作者：
De Graef, M.
中科院分区：
文献类型：
--
作者：
Ding, Z.;Pascal, E.;De Graef, M.
EBSD模式的准确索引提出了一个具有挑战性的问题。我们提出了一种新的卷积神经网络（EBSD-CNN）来实现EBSD模式的实时索引;我们实现了一个失向损失函数来适应标准CNN模型用于晶体学取向索引。使用模拟和实验数据的索引的准确性，率，和对噪声的鲁棒性进行评估，并与其他索引方法（基于Hough的索引，字典索引，球形索引）进行比较。结果表明，CNN可以提供商业Hough变换索引的替代方案，具有相当的准确性和速度。我们通过可视化选定的过滤器来深入了解网络功能。(C)2020 Acta Materialia Inc.爱思唯尔有限公司出版
Accurate indexing of EBSD patterns presents a challenging problem. We propose a new convolutional neural network (EBSD-CNN) to realize real-time indexing of EBSD patterns; we implement a disorientation loss function to adapt a standard CNN model for crystallographic orientation indexing. The indexing accuracy, rate, and robustness against noise are evaluated using both simulated and experimental data, and compared with other indexing methods (Hough-based indexing, dictionary indexing, and spherical indexing). The results suggest that a CNN can provide an alternative to commercial Hough-transform-based indexing with comparative accuracy and rate. We obtain insight into the network functionality by visualization of selected filters. (C) 2020 Acta Materialia Inc. Published by Elsevier Ltd.