Deep learning analysis of polaritonic waves images

TL;DR

This paper demonstrates a deep learning approach using CNNs to rapidly analyze nano-scale polaritonic wave images, extracting key parameters much faster than traditional methods, validated on experimental data.

Contribution

Developed a CNN-based protocol for fast, accurate analysis of polaritonic images, enabling simultaneous extraction of wavelength, quality factor, and material parameters.

Findings

CNN analysis is at least 1000 times faster than traditional fitting methods.

Validated approach on experimental images of charge-transfer plasmon polaritons.

Framework applicable to various scanning probe imaging techniques.

Abstract

Deep learning (DL) is an emerging analysis tool across sciences and engineering. Encouraged by the successes of DL in revealing quantitative trends in massive imaging data, we applied this approach to nano-scale deeply sub-diffractional images of propagating polaritonic waves in complex materials. We developed a practical protocol for the rapid regression of images that quantifies the wavelength and the quality factor of polaritonic waves utilizing the convolutional neural network (CNN). Using simulated near-field images as training data, the CNN can be made to simultaneously extract polaritonic characteristics and materials parameters in a timescale that is at least three orders of magnitude faster than common fitting/processing procedures. The CNN-based analysis was validated by examining the experimental near-field images of charge-transfer plasmon polaritons at Graphene/{\alpha}-RuCl3 interfaces. Our work provides a general framework for extracting quantitative information from images generated with a variety of scanning probe methods.