3D Deep Learning Enables Accurate Layer Mapping of 2D Materials
Xingchen Dong,Hongwei Li,Zhutong Jiang,Theresa Grünleitner,İnci Güler,Jie Dong,Kun Wang,Michael H. Köhler,Martin Jakobi,Bjoern H. Menze,Ali K. Yetisen,Ian D. Sharp,Andreas V. Stier,Jonathan J. Finley,Alexander W. Koch,Theresa Grünleitner,İnci Güler,Michael H. Köhler
DOI: https://doi.org/10.1021/acsnano.0c09685
IF: 17.1
2021-01-19
ACS Nano
Abstract:Layered, two-dimensional (2D) materials are promising for next-generation photonics devices. Typically, the thickness of mechanically cleaved flakes and chemical vapor deposited thin films is distributed randomly over a large area, where accurate identification of atomic layer numbers is time-consuming. Hyperspectral imaging microscopy yields spectral information that can be used to distinguish the spectral differences of varying thickness specimens. However, its spatial resolution is relatively low due to the spectral imaging nature. In this work, we present a 3D deep learning solution called DALM (deep-learning-enabled atomic layer mapping) to merge hyperspectral reflection images (high spectral resolution) and RGB images (high spatial resolution) for the identification and segmentation of MoS<sub>2</sub> flakes with mono-, bi-, tri-, and multilayer thicknesses. DALM is trained on a small set of labeled images, automatically predicts layer distributions and segments individual layers with high accuracy, and shows robustness to illumination and contrast variations. Further, we show its advantageous performance over the state-of-the-art model that is solely based on RGB microscope images. This AI-supported technique with high speed, spatial resolution, and accuracy allows for reliable computer-aided identification of atomically thin materials.The Supporting Information is available free of charge at <a class="ext-link" href="/doi/10.1021/acsnano.0c09685?goto=supporting-info">https://pubs.acs.org/doi/10.1021/acsnano.0c09685</a>.Photographs of the optical microscope and the hyperspectral imaging microscope for data acquisition; GUI for dual-modality image registration and layer subclass labeling; labeled data and pixels with different values showing different layer numbers; principle of image conversion from labeled data to one-hot images of different layer numbers; data dimension variation from raw data to data pairs suitable for network training; data augmentation of RGB images; dimension fusion strategy of DALM; optical RGB image, normalized hyperspectral images, and normalized spectra of one data pair for demonstration; training curves of both models; confusion matrix of the network prediction using a rough classification of different subclasses; component list of the hyperspectral imaging microscope (<a class="ext-link" href="/doi/suppl/10.1021/acsnano.0c09685/suppl_file/nn0c09685_si_001.pdf">PDF</a>)This article has not yet been cited by other publications.
materials science, multidisciplinary,chemistry, physical,nanoscience & nanotechnology