Training Neural Networks to Classify Chiral Nanoparticle Stereopairs Using Weakly Labelled Datasets

Alexander J Pattison, Mary Scott, Peter Ercius, Wolfgang Theis; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2024, pp. 8338-8342

Abstract

As the interest in using machine learning techniques for electron microscopy grows so does the need for labeled datasets and automated labeling strategies. Here we exploit the nature of chirality to weakly label datasets of scanning transmission electron microscope stereopairs of chiral tellurium nanoparticles. We then use these datasets to train convolutional neural networks to classify the handedness of these particles. The methods and results can be applied to other machine learning methods involving weak labeling and stereo imaging.

Related Material

[pdf]
[bibtex]
@InProceedings{Pattison_2024_CVPR, author = {Pattison, Alexander J and Scott, Mary and Ercius, Peter and Theis, Wolfgang}, title = {Training Neural Networks to Classify Chiral Nanoparticle Stereopairs Using Weakly Labelled Datasets}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2024}, pages = {8338-8342} }