Intelligent Identification of MoS2Nanostructures with Hyperspectral Imaging by 3D-CNN
Intelligent Identification of MoS2Nanostructures with Hyperspectral Imaging by 3D-CNN
复制标题
DOI：
10.3390/nano10061161
复制
发表时间：
2020-06-01
期刊：
影响因子：
5.3
通讯作者：
Wang, Hsiang-Chen
中科院分区：
文献类型：
--
作者：
Li, Kai-Chun;Lu, Ming-Yen;Wang, Hsiang-Chen
二维材料由于其优异的性能和晶圆级合成方法而受到越来越多的关注。然而，二维材料纳米结构的大面积表征、精密、智能自动化、高效检测尚未达到工业水平。因此，我们利用大数据分析和深度学习的方法，成功开发了一套适用于少层二硫化钼自动识别的可见光高光谱成像技术。在分类算法方面，我们提出了深度神经网络、一维（1D）卷积神经网络和三维（3D）卷积神经网络（3D- cnn）模型，探索模型识别精度与少层MoS2光学特性之间的相关性。实验结果表明，3D-CNN比其他分类模型具有更好的泛化能力，该模型适用于空间域和光谱域的特征输入。这种差异存在于没有特定衬底的本研究的先前版本中，并且可以通过自动快门光圈管理样品部分上不同动态范围的图像。因此，在相同颜色对比度条件下，无需调整成像质量，也无需采用常规图像的处理方法来实现类似1.92 mm(2)的最大视场识别范围。图像分辨率可达到接近100 nm，每张图像检测时间为3 min。
Increasing attention has been paid to two-dimensional (2D) materials because of their superior performance and wafer-level synthesis methods. However, the large-area characterization, precision, intelligent automation, and high-efficiency detection of nanostructures for 2D materials have not yet reached an industrial level. Therefore, we use big data analysis and deep learning methods to develop a set of visible-light hyperspectral imaging technologies successfully for the automatic identification of few-layers MoS2. For the classification algorithm, we propose deep neural network, one-dimensional (1D) convolutional neural network, and three-dimensional (3D) convolutional neural network (3D-CNN) models to explore the correlation between the accuracy of model recognition and the optical characteristics of few-layers MoS2. The experimental results show that the 3D-CNN has better generalization capability than other classification models, and this model is applicable to the feature input of the spatial and spectral domains. Such a difference consists in previous versions of the present study without specific substrate, and images of different dynamic ranges on a section of the sample may be administered via the automatic shutter aperture. Therefore, adjusting the imaging quality under the same color contrast conditions is unnecessary, and the process of the conventional image is not used to achieve the maximum field of view recognition range of similar to 1.92 mm(2). The image resolution can reach similar to 100 nm and the detection time is 3 min per one image.