Pixel-wise classification in graphene-detection with tree-based machine learning algorithms

TL;DR

This paper presents supervised pixel-wise classification methods using tree-based machine learning algorithms to automate graphene layer identification from optical microscopy images, achieving high performance with minimal training data and computational resources.

Contribution

Introduction of four tree-based machine learning algorithms for pixel-wise graphene classification, including evaluation of their performance and combinatorial models.

Findings

High classification accuracy with small datasets

Effective performance without GPU computation

Open-source code for community use

Abstract

Mechanical exfoliation of graphene and its identification by optical inspection is one of the milestones in condensed matter physics that sparked the field of 2D materials. Finding regions of interest from the entire sample space and identification of layer number is a routine task potentially amenable to automatization. We propose supervised pixel-wise classification methods showing a high performance even with a small number of training image datasets that require short computational time without GPU. We introduce four different tree-based machine learning algorithms -- decision tree, random forest, extreme gradient boost, and light gradient boosting machine. We train them with five optical microscopy images of graphene, and evaluate their performances with multiple metrics and indices. We also discuss combinatorial machine learning models between the three single classifiers and assess their performances in identification and reliability. The code developed in this paper is open to the public and will be released at github.com/gjung-group/Graphene_segmentation.