Data and supplementary material for Improving U-Net Confidence on TEM Image Data with L2-Regularization, Transfer Learning, and Deep Fine-Tuning

With ever-increasing data volumes, it is essential to develop automated approaches for identifying nanoscale defects in transmission electron microscopy (TEM) images. However, compared to features in conventional photographs, nanoscale defects in TEM images exhibit far greater variation due to the complex contrast mechanisms and intricate
defect structures. These challenges often result in much less
labelled data and higher rates of annotation errors, posing
significant obstacles to improving machine learning model
performance for TEM image analysis. To address these limitations, we examined transferring learning by leveraging
large, pre-trained models used for natural images.

We demonstrated that by using the pre-trained encoder
and L2-regularization, semantically complex features are
ignored in favor of simpler, more reliable cues, substantially
improving the model performance. However, this improvement cannot be captured by conventional evaluation metrics
such as F1-score, which can be skewed by human annotation
errors treated as “ground truth”. Instead, we introduced
novel evaluation metrics that are independent of the annotation accuracy. Using grain boundary detection in UO2 as a
case study, we found that our approach led to a 64% increase
in the total number of grains detected, which acts a robust
and holistic measure of model performance on the TEM
dataset used in this work. Finally, we showed that model
self-confidence is only achieved through transfer learning
and fine-tuning of very deep layers.