Abstract

The inverse scattering problem of non-spherical particle size estimation is solved using a series of supervised machine learning models trained on a library of light scattering data. By establishing a large library with spheres and spheroids as fundamental shapes and through optimization of model hyperparameters, the trained models are able to accurately estimate a precise equivalent volume sphere radius of particles from an external database and simulations, with root mean square errors of 2.6% and 1.9% for the external and simulated particles, respectively. It was found that classification via a $ k $-nearest neighbor model and refinement via a trained ensemble regression model performed best for equivalent volume measurements.

© 2020 Optical Society of America

Machine-learning-based computationally efficient particle size distribution retrieval from bulk optical properties

Ruhui Jia, Xiaohao Zhang, Fenping Cui, Gongye Chen, Haomiao Li, Haochen Peng, Zhaolou Cao, and Shixin Pei
Appl. Opt. 59(24) 7284-7291 (2020)

Use of light scattering to estimate the fraction of spherical particles in a mixture

Paul Rochon, T. J. Racey, and M. Zeller
Appl. Opt. 27(15) 3295-3298 (1988)

Machine Learning Models for Estimating Quality of Transmission in DWDM Networks

Rui Manuel Morais and João Pedro
J. Opt. Commun. Netw. 10(10) D84-D99 (2018)

Size distribution of mineral aerosol: using light-scattering models in laser particle sizing

Ben Veihelmann, Martin Konert, and Wim J. van der Zande
Appl. Opt. 45(23) 6022-6029 (2006)

Machine learning holography for 3D particle field imaging

Siyao Shao, Kevin Mallery, S. Santosh Kumar, and Jiarong Hong
Opt. Express 28(3) 2987-2999 (2020)