Application of a deep learning semantic segmentation model to helium bubbles and voids in nuclear materials

Agarwal, Shradha; Sawant, Anshul; Faisal, Muhammad; Copp, Sydney E.; Reyes-Zacarias, July; Lin, Yan-Ru; Zinkle, Steven J.

doi:10.1016/j.engappai.2023.106747

Application of a deep learning semantic segmentation model to helium bubbles and voids in nuclear materials

Journal Article · Tue Aug 01 00:00:00 EDT 2023 · Engineering Applications of Artificial Intelligence

DOI:https://doi.org/10.1016/j.engappai.2023.106747· OSTI ID:2229233

Agarwal, Shradha ^[1]; ^[2]; ^[3]; ^[1]; Reyes-Zacarias, July ^[1]; ^[4]; Zinkle, Steven J. ^[5]

Univ. of Tennessee, Knoxville, TN (United States)
Google, Sunnyvale, CA (United States)
Saudi Network Services, Riyadh (Saudi Arabia)
Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
Univ. of Tennessee, Knoxville, TN (United States); Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Imaging nanoscale radiation-induced defects using the transmission electron microscope (TEM) is a key factor in the successful implementation of materials for nuclear energy structural applications. Analyzing each defect in a TEM micrograph is currently a manual task. To identify the defects in a single image can take anywhere from 15 min to an hour and a project can require the analysis of anywhere from tens to ≥ 100 images. Here, we use artificial intelligence (AI) models to automate this task. For simplification, we evaluated images with only a single type of defect; helium bubbles. Additionally, we performed semantic segmentation of these helium bubble defects in electron microscopy images of irradiated FeCrAl alloys using a deep learning DefectSegNet model. This model, which was previously used to classify crystal defects, is inspired by the classic DenseNet and U-Net image segmentation models. It claims high spatial resolution, but has poor performance at object boundaries. Our paper improves the DefectSegNet model’s application by adding two new features. First, the DefectSegNet model is applied not only to perform calculation pixel-wise but also object (or feature) wise. Because object-wise metrics are directly relevant to our final goal of detecting bubbles, whereas pixel-wise classification is only an intermediate step, it is an important part of our study. Second, a distance map loss (DML) function has been added to increase its performance at object boundaries. It is crucial to accurately represent defects boundaries, especially bubbles, in order to track the bubble-induced swelling caused by irradiation. The boundary-focused DML function is also compared to other loss functions like Cross-entropy, Weighted Binary Cross Entropy (WBCE), Dice and Intersection over Union (IOU). Finally, by incorporating new features, we found a marked improvement on segmentation quality and better shape preservation at the boundaries and areas of the bubbles.

View Accepted Manuscript (DOE)

Research Organization:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: USDOE

Grant/Contract Number:: AC05-00OR22725

OSTI ID:: 2229233

Journal Information:: Engineering Applications of Artificial Intelligence, Journal Name: Engineering Applications of Artificial Intelligence Journal Issue: Part A Vol. 126; ISSN 0952-1976

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (16)

U-Net: Convolutional Networks for Biomedical Image Segmentation Ronneberger, Olaf; Fischer, Philipp; Brox, Thomas Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III https://doi.org/10.1007/978-3-319-24574-4_28	book	November 2015
Effect of mobile helium on void nucleation in materials during irradiation Wiedersich, Hartmut; Burton, J. J.; Katz, Joseph L. Journal of Nuclear Materials, Vol. 51, Issue 3 https://doi.org/10.1016/0022-3115(74)90194-9	journal	July 1974
Temperature-dependent cavity swelling in dual-ion irradiated Fe and Fe-Cr ferritic alloys Lin, Yan-Ru; Bhattacharya, Arunodaya; Chen, Da Acta Materialia, Vol. 207 https://doi.org/10.1016/j.actamat.2021.116660	journal	April 2021
Bubble formation in helium-implanted nanostructured ferritic alloys at elevated temperatures Lin, Yan-Ru; Chen, Wei-Ying; Tan, Lizhen Acta Materialia, Vol. 217 https://doi.org/10.1016/j.actamat.2021.117165	journal	September 2021
Deep learning object detection in materials science: Current state and future directions Jacobs, Ryan Computational Materials Science, Vol. 211 https://doi.org/10.1016/j.commatsci.2022.111527	journal	August 2022
A review on helium mobility in inorganic materials Trocellier, P.; Agarwal, S.; Miro, S. Journal of Nuclear Materials, Vol. 445, Issue 1-3 https://doi.org/10.1016/j.jnucmat.2013.10.061	journal	February 2014
Helium mobility in advanced nuclear ceramics Agarwal, S.; Trocellier, P.; Serruys, Y. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, Vol. 327 https://doi.org/10.1016/j.nimb.2013.09.039	journal	May 2014
Automated Detection of Helium Bubbles in Irradiated X-750 Anderson, Chris M.; Klein, Jacob; Rajakumar, Heygaan Ultramicroscopy, Vol. 217 https://doi.org/10.1016/j.ultramic.2020.113068	journal	October 2020
Deep Learning for Semantic Segmentation of Defects in Advanced STEM Images of Steels Roberts, Graham; Haile, Simon Y.; Sainju, Rajat Scientific Reports, Vol. 9, Issue 1 https://doi.org/10.1038/s41598-019-49105-0	journal	September 2019
The influence of helium on the bulk properties of fusion reactor structural materials Ullmaier, H. Nuclear Fusion, Vol. 24, Issue 8 https://doi.org/10.1088/0029-5515/24/8/009	journal	August 1984
Fully convolutional networks for semantic segmentation Long, Jonathan; Shelhamer, Evan; Darrell, Trevor 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2015.7298965	conference	June 2015
Densely Connected Convolutional Networks Huang, Gao; Liu, Zhuang; Maaten, Laurens van der 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR.2017.243	conference	July 2017
Rethinking Semantic Segmentation: A Prototype View Zhou, Tianfei; Wang, Wenguan; Konukoglu, Ender 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR52688.2022.00261	conference	June 2022
Learning Deconvolution Network for Semantic Segmentation Noh, Hyeonwoo; Hong, Seunghoon; Han, Bohyung 2015 IEEE International Conference on Computer Vision (ICCV) https://doi.org/10.1109/ICCV.2015.178	conference	December 2015
SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation Badrinarayanan, Vijay; Kendall, Alex; Cipolla, Roberto IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 39, Issue 12 https://doi.org/10.1109/TPAMI.2016.2644615	journal	December 2017
Radiation-Induced Helium Bubbles in Metals Li, Shi-Hao; Li, Jing-Ting; Han, Wei-Zhong Materials, Vol. 12, Issue 7 https://doi.org/10.3390/ma12071036	journal	March 2019