Rapid identification of two-dimensional materials via machine learning assisted optic microscopy
Rapid identification of two-dimensional materials via machine learning assisted optic microscopy
复制标题
通过机器学习辅助光学显微镜快速识别二维材料
DOI：
10.1016/j.jmat.2019.03.003
复制
发表时间：
2019-09-01
影响因子：
9.4
通讯作者：
Wu, Rong
中科院分区：
文献类型：
--
作者：
Li, Yuhao;Kong, Yangyang;Wu, Rong
提出了一种结合菲涅耳定律和机器学习的方法来识别二维材料的层数。提出了光学对比度、红绿蓝、总色差3个指标来描述和模拟Si/SiO2衬底上二维材料的可见性，采用k均值聚类和k近邻机器学习算法获得二维材料的厚度数据库，并通过红绿蓝指标对二维材料的光学图像进行检验。结果表明，该方法可以快速、准确、大面积地提供二维材料的特性。随着人工智能和纳米科学的结合，这种机器学习辅助方法减轻了工作量，促进了2D材料的基础研究。(C)2019中国陶瓷学会制作和主办：Elsevier B. V.
A combination of Fresnel law and machine learning method is proposed to identify the layer counts of 2D materials. Three indexes, which are optical contrast, red-green-blue, total color difference, are presented to illustrate and simulate the visibility of 2D materials on Si/SiO2 substrate, and the machine learning algorithms, which are k-mean clustering and k-nearest neighbors, are employed to obtain thickness database of 2D material and test the optical images of 2D materials via red-green-blue index. The results show that this method can provide fast, accurate and large-area property of 2D material. With the combination of artificial intelligence and nanoscience, this machine learning assisted method eases the workload and promotes fundamental research of 2D materials. (C) 2019 The Chinese Ceramic Society. Production and hosting by Elsevier B.V.