: Point defects play a fundamental role in the discovery of new materials due to their strong influence on material properties and behavior. At present, imaging techniques based on transmission electron microscopy (TEM) are widely employed for characterizing point defects in materials. However, current methods for defect detection predominantly involve visual inspection of TEM images. Recent efforts to develop machine learning methods for the detection of point defects in TEM images have focused on supervised methods that require labeled training data that is generated via simulation. Motivated by a desire for machine learning methods that can be trained on experimental data, we propose a self-supervised machine learning algorithm that is trained solely on images that are defect-free. Our proposed method uses principal components analysis (PCA) and a convolutional neural network (CNN) to analyze a TEM image and predict the location of a defect. Using simulated TEM images, we show that the PCA-CNN model is able to accurately locate point defects even with significant levels of imaging noise. We also apply the PCA-CNN model to an experimental TEM image and show that it outperforms state-of-the-art, general purpose defect detection models.
Copyright. All Rights Reserved.