Identification of Two-Dimensional Material Nanosheets Based on Deep Neural Network and Hyperspectral Microscopy Images
PENG Ren-miao1, 2, XU Peng-peng2, ZHAO Yi-mo2, BAO Li-jun1, LI Cheng2*
1. School of Electronic Science and Engineering，Xiamen University, Xiamen 361005, China
2. College of Physical Science and Technology, Xiamen University, Xiamen 361005, China
Abstract：In recent years, two-dimensional materials have received widespread attention due to their unique properties. Among the various methods for preparing two-dimensional layered crystals, the thin-layer two-dimensional material crystals obtained by mechanical exfoliation are of high quality, suitable for basic research and performance demonstration. However, mechanically exfoliated crystals on substrates exhibit a certain degree of randomness, including a few layers and relatively thick flakes. The effective, rapid and intelligent characterization method of these two-dimensional nanostructures is beneficial to further research on the properties of two-dimensional materials. This paper proposes a method based on deep learning, which can segment and quickly identify two-dimensional material nanosheets based on optical microscope images through a convolutional neural network semantic segmentation algorithm built with an encoding-decoding structure. As a typical algorithm for deep learning in the field of image processing, Convolutional neural networks can be applied to the feature extraction in optical microscope images. Firstly, MoS2 nanosheet samples were prepared by mechanical exfoliation, and high spectroscopic images were acquired by optical microscopy. The nanosheet samples were labeled, and the marked images were further processed, including color calibration and sliding shear operation, to obtain datasets for network training and testing. A semantic segmentation algorithm based on encoding-decoding network structure was designed to rapidly identify nanosheets. Aiming at some flakes in images showing the characteristic of low contrast and fragmentation, residual convolution and pyramid pooling models were added to strengthen the extraction of features during encoding. The shallow feature information extracted from the encoding stage was reused during decoding to improve the network segmentation results. In the experiment, the weighted cross-entropy loss function was used to solve the problem of unbalanced classes, and the dataset was enlarged with data augmentation. Testing on the trained network show that the pixel accuracy was 97.38%, the mean pixel accuracy was 90.38%, and the mean intersection over union was 75.86%. Then, the exfoliated monolayer and bilayer graphenes were identified by transfer learning, and the mean intersection over union reached 81.63%, showing that this technique is universal for the identification of two-dimensional nanosheets. The identification of MoS2 and graphene nanosheets realizes the application of deep learning in optical microscopy images of two-dimensional materials. This method is expected to apply to more two-dimensional materials and break through the problem of automatic dynamic processing of optical images. Moreover, it provides a reference for hyperspectral images processing of other nanomaterials.
彭仁苗，徐鹏鹏，赵一默，包立君，李 成. 深度神经网络和高光谱显微图像的二维材料纳米片识别[J]. 光谱学与光谱分析, 2022, 42(06): 1965-1973.
PENG Ren-miao, XU Peng-peng, ZHAO Yi-mo, BAO Li-jun, LI Cheng. Identification of Two-Dimensional Material Nanosheets Based on Deep Neural Network and Hyperspectral Microscopy Images. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1965-1973.
[1] Novoselov K S, Geim A K, Morozov S V, et al. Science, 2004, 306(5296): 666.
[2] Wang Q H, Kalantar-Zadeh K, Kis A, et al. Nature Nanotechnology, 2012, 7(11): 699.
[3] Buscema M, Groenendijk D J, Blanter S I, et al. Nano Letters, 2014, 14(6): 3347.
[4] Zhang K, Feng Y, Wang F, et al. Journal of Materials Chemistry C, 2017, 5(46): 11992.
[5] Li H, Wu J, Huang X, et al. ACS Nano, 2013, 7(11): 10344.
[6] WANG Can, WU Xin-hui, LI Lian-qing, et al(王 璨, 武新慧, 李恋卿, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2018, 38(1): 36.
[7] Krizhevsky A, Sutskever I, Hinton G E, et al. Communications of the ACM, 2017, 60(6): 84.
[8] Long J, Shelhamer E, Darrell T. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2015, 39(4): 640.
[9] Ronneberger O, Fischer P, Brox T. International Conference on Medical Image Computing and Computer-Assisted Intervention, 2015, 9351: 234.
[10] Badrinarayanan V, Kendall A, Cipolla R. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017, 39(12): 2481.
[11] Grvik E, Yi D, Iv M, et al. Journal of Magnetic Resonance Imaging, 2020, 51(1): 175.
[12] Wang C, Zhao Z, Ren Q, et al. Entropy, 2019, 21(2): 168.
[13] He K, Zhang X, Ren S, et al. Deep Residual Learning for Image Recognition, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.
[14] Zhao H S, Shi J P, Qi X J, et al. Pyramid Scene Parsing Network, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
[15] Lateef F, Ruichek Y. Neurocomputing, 2019, 338: 321.
