Automated Classification of Helium Ingress in Irradiated X-750

TL;DR

This paper presents an automated deep learning approach for detecting helium bubbles in micrographs of irradiated Inconel X-750, achieving human-level accuracy and significantly faster analysis for nuclear materials research.

Contribution

It adapts a region-based convolutional neural network for helium bubble detection, enabling rapid, accurate, and reproducible micrograph analysis across various imaging conditions.

Findings

Neural network achieves accuracy comparable to human analysis.

Analysis speed is four orders of magnitude faster than manual methods.

Method works across different micrograph contrasts and magnifications.

Abstract

Imaging nanoscale features using transmission electron microscopy is key to predicting and assessing the mechanical behavior of structural materials in nuclear reactors. Analyzing these micrographs is often a tedious and labour intensive manual process. It is a prime candidate for automation. Here, a region-based convolutional neural network is adapted to detect helium bubbles in micrographs of neutron-irradiated Inconel X-750 reactor spacer springs. We demonstrate that this neural network produces analyses of similar accuracy and reproducibility to that produced by humans. Further, we show this method as being four orders of magnitude faster than manual analysis allowing for generation of significant quantities of data. The proposed method can be used with micrographs of different Fresnel contrasts and magnification levels.