A Deep Learning Approach for Semantic Segmentation of Unbalanced Data in Electron Tomography of Catalytic Materials

TL;DR

This paper presents a deep learning method using U-Net with weighted focal loss for accurate semantic segmentation of catalytic materials in electron tomography, effectively handling class imbalance and complex surface structures.

Contribution

It introduces a novel application of weighted focal loss with U-Net for segmentation of unbalanced electron microscopy data, achieving high accuracy in complex catalytic materials.

Findings

Achieved an average DSC of 0.96 for support material segmentation.

Achieved an average DSC of 0.84 for Pt nanoparticle segmentation.

Boundary-overlap error less than 2 nm at the 90th percentile of Hausdorff distance.

Abstract

Heterogeneous catalysts possess complex surface and bulk structures, relatively poor intrinsic contrast, and often a sparse distribution of the catalytic nanoparticles (NPs), posing a significant challenge for image segmentation, including the current state-of-the-art deep learning methods. To tackle this problem, we apply a deep learning-based approach for the multi-class semantic segmentation of a $\gamma$-Alumina/Pt catalytic material in a class imbalance situation. Specifically, we used the weighted focal loss as a loss function and attached it to the U-Net's fully convolutional network architecture. We assessed the accuracy of our results using Dice similarity coefficient (DSC), recall, precision, and Hausdorff distance (HD) metrics on the overlap between the ground-truth and predicted segmentations. Our adopted U-Net model with the weighted focal loss function achieved an average DSC score of 0.96 $\pm$ 0.003 in the $\gamma$-Alumina support material and 0.84 $\pm$ 0.03 in the Pt NPs segmentation tasks. We report an average boundary-overlap error of less than 2 nm at the 90th percentile of HD for $\gamma$-Alumina and Pt NPs segmentations. The complex surface morphology of the $\gamma$-Alumina and its relation to the Pt NPs were visualized in 3D by the deep learning-assisted automatic segmentation of a large data set of high-angle annular dark-field (HAADF) scanning transmission electron microscopy (STEM) tomography reconstructions.