Deep‐Learning‐Based Microscopic Imagery Classification, Segmentation, and Detection for the Identification of 2D Semiconductors. Issue 9 (13th July 2022)

• Record Type: Journal Article
• Authors: Dong, Xingchen; Li, Hongwei; Yan, Yuntian; Cheng, Haoran; Zhang, Hui Xin; Zhang, Yucheng; Le, Tien Dat; Wang, Kun; Dong, Jie; Jakobi, Martin; Yetisen, Ali K.; Koch, Alexander W.
• Is Part Of: Advanced theory and simulations. Volume 5:Issue 9(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-07-13
• Subjects: 2D semiconductors -- classification -- detection -- large‐area identification -- machine learning -- microscopic imagery -- segmentation; Science -- Simulation methods -- Periodicals; Science -- Methodology -- Periodicals; Engineering -- Simulation methods -- Periodicals; Engineering -- Methodology -- Periodicals; 507.21
• DOI: 10.1002/adts.202200140
• Languages: English
• ISSNs: 2513-0390

Abstract

Abstract: 2D materials and their heterostructures are prominent for fabricating next‐generation optical and photonic devices. The optical, electrical, and mechanical properties of 2D materials largely depend on atomic layer numbers. Although machine learning techniques are implemented to identify large‐area thickness distribution using microscopic images, the existing work mainly focuses on rough identification of thicknesses with in‐house datasets which limits fair and comprehensive comparisons of new machine learning approaches. Here, first a microscopic dataset is collected and released for three fundamental image processing tasks including multilabel classification, segmentation, and detection. Then three deep‐learning architectures DenseNet, U‐Net, and Mask‐region convolutional neural network (RCNN) are benchmarked on three tasks and their robustness is evaluated on the augmented 2D microscopic images with different optical contrast variations. Deep learning models are trained and evaluated to identify mono‐, bi‐, tri‐, multilayer and bulk flakes using microscopic images of MoS2 fabricated on the SiO2 /Si substrate by chemical vapor deposition. The relation between model performances and statistics of datasets is studied based on the international commission on illumination (CIE) 1931 color space and red, green, blue (RGB) histograms of optical contrast differences. Finally, the robust pretrained models are integrated into a graphic user interface for the on‐site use of full field‐of‐view images captured by bright‐field microscopes. Abstract : To identify large‐area thickness distribution of 2D materials using microscopic images, deep‐learning architectures are used to do multilabel classification, segmentation, and detection tasks. The relation between model performances and dataset statistics is analyzed based on the color space analysis and optical contrast differences. The robust pretrained models are implemented into a graphic user interface for the on‐site use.