Automatic detection of multilayer hexagonal boron nitride in optical images using deep learning-based computer vision

• Published in: Scientific reports Vol. 13; no. 1; p. 1595
• Main Authors: Ramezani, Fereshteh, Parvez, Sheikh, Fix, J. Pierce, Battaglin, Arthur, Whyte, Seamus, Borys, Nicholas J., Whitaker, Bradley M.
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 28.01.2023; Nature Publishing Group; Nature Portfolio
• Subjects: 639/301/357/1018; 639/705/794; Algorithms; Boron; Computer vision; Crystals; Deep learning; Exfoliation; Humanities and Social Sciences; Light microscopy; Micrography; multidisciplinary; Science; Science (multidisciplinary)
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-023-28664-3

Computer vision algorithms can quickly analyze numerous images and identify useful information with high accuracy. Recently, computer vision has been used to identify 2D materials in microscope images. 2D materials have important fundamental properties allowing for their use in many potential applications, including many in quantum information science and engineering. One such material is hexagonal boron nitride (hBN), an isomorph of graphene with a very indistinguishable layered structure. In order to use these materials for research and product development, the most effective method is mechanical exfoliation where single-layer 2D crystallites must be prepared through an exfoliation procedure and then identified using reflected light optical microscopy. Performing these searches manually is a time-consuming and tedious task. Deploying deep learning-based computer vision algorithms for 2D material search can automate the flake detection task with minimal need for human intervention. In this work, we have implemented a new deep learning pipeline to classify crystallites of hBN based on coarse thickness classifications in reflected-light optical micrographs. We have used DetectoRS as the object detector and trained it on 177 images containing hexagonal boron nitride (hBN) flakes of varying thickness. The trained model achieved a high detection accuracy for the rare category of thin flakes ( < 50 atomic layers thick). Further analysis shows that our proposed pipeline could be generalized to various microscope settings and is robust against changes in color or substrate background.