Semantic Segmentation of Micrographs for Nuclear Fuel Analysis and Degradation Quantification

Christopher Snyder; Katherine I. Montoya; David Patrick; Jordan Stone; Daniel Mohanadhas; Elizabeth S. Sooby; Amanda S. Fernandez

doi:10.1109/ACCESS.2023.3326074

Semantic Segmentation of Micrographs for Nuclear Fuel Analysis and Degradation Quantification

Christopher Snyder, Katherine I. Montoya, David Patrick, Jordan Stone, Daniel Mohanadhas, Elizabeth S. Sooby, Amanda S. Fernandez

Research output: Contribution to journal › Article › peer-review

Abstract

When fuel materials for high-temperature gas-cooled nuclear reactors are quantification tested, significant analysis is required to establish their stability under various proposed accident scenarios, as well as to assess degradation over time. Typically, samples are examined by lab assistants trained to capture micrograph images used to analyze the degradation of a material. Analysis of these micrographs still require manual intervention which is time-consuming and can introduce human-error. While machine learning and computer vision models would be useful to this analysis, data for training such models is limited due to physical experiment costs, including lab hours and materials. This collaborative research are: 1) establishes an open dataset of micrographs and semantic labels named Graphite-23; 2) analyzes semantic segmentation architectures against the new data; and 3) contributes open source code for the community to progress research in degradation analysis of materials. A U-Net architecture with various backbones demonstrates competitive performance on the proposed dataset, with an mIoU up to 0.83, establishing a clear baseline for future research in this intersection of fields.

Original language	English
Pages (from-to)	118512-118520
Number of pages	9
Journal	IEEE Access
Volume	11
DOIs	https://doi.org/10.1109/ACCESS.2023.3326074
State	Published - 2023
Externally published	Yes

Funding

This work was supported in part by the Department of Energy (National Nuclear Security Administration Minority Serving Institution Partnership Program's CONNECT-the CONsortium on Nuclear sECurity Technologies) under Grant DE-NA0003948 and Grant DE-NA0004107, and in part by the U.S. Department of Energy Idaho National Laboratory (DOE INL) Nuclear Energy University Program under Project 21-24522. This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

Funders	Funder number
DOE INL	21-24522
National Nuclear Security Administration Minority Serving Institution	DE-NA0004107, DE-NA0003948
U.S. Department of Energy Idaho National Laboratory
United States Government
U.S. Department of Energy

Keywords

Computer vision
deep learning
fuel analysis
graphite
neural networks
nuclear materials
semantic segmentation

Access to Document

10.1109/ACCESS.2023.3326074

Cite this

@article{3dcc1ca5f0fc4fc1a6abfce87dde7508,

title = "Semantic Segmentation of Micrographs for Nuclear Fuel Analysis and Degradation Quantification",

abstract = "When fuel materials for high-temperature gas-cooled nuclear reactors are quantification tested, significant analysis is required to establish their stability under various proposed accident scenarios, as well as to assess degradation over time. Typically, samples are examined by lab assistants trained to capture micrograph images used to analyze the degradation of a material. Analysis of these micrographs still require manual intervention which is time-consuming and can introduce human-error. While machine learning and computer vision models would be useful to this analysis, data for training such models is limited due to physical experiment costs, including lab hours and materials. This collaborative research are: 1) establishes an open dataset of micrographs and semantic labels named Graphite-23; 2) analyzes semantic segmentation architectures against the new data; and 3) contributes open source code for the community to progress research in degradation analysis of materials. A U-Net architecture with various backbones demonstrates competitive performance on the proposed dataset, with an mIoU up to 0.83, establishing a clear baseline for future research in this intersection of fields.",

keywords = "Computer vision, deep learning, fuel analysis, graphite, neural networks, nuclear materials, semantic segmentation",

author = "Christopher Snyder and Montoya, {Katherine I.} and David Patrick and Jordan Stone and Daniel Mohanadhas and Sooby, {Elizabeth S.} and Fernandez, {Amanda S.}",

note = "Publisher Copyright: {\textcopyright} 2023 The Authors.",

year = "2023",

doi = "10.1109/ACCESS.2023.3326074",

language = "English",

volume = "11",

pages = "118512--118520",

journal = "IEEE Access",

issn = "2169-3536",

publisher = "Institute of Electrical and Electronics Engineers",

}

TY - JOUR

T1 - Semantic Segmentation of Micrographs for Nuclear Fuel Analysis and Degradation Quantification

AU - Snyder, Christopher

AU - Montoya, Katherine I.

AU - Patrick, David

AU - Stone, Jordan

AU - Mohanadhas, Daniel

AU - Sooby, Elizabeth S.

AU - Fernandez, Amanda S.

PY - 2023

Y1 - 2023

N2 - When fuel materials for high-temperature gas-cooled nuclear reactors are quantification tested, significant analysis is required to establish their stability under various proposed accident scenarios, as well as to assess degradation over time. Typically, samples are examined by lab assistants trained to capture micrograph images used to analyze the degradation of a material. Analysis of these micrographs still require manual intervention which is time-consuming and can introduce human-error. While machine learning and computer vision models would be useful to this analysis, data for training such models is limited due to physical experiment costs, including lab hours and materials. This collaborative research are: 1) establishes an open dataset of micrographs and semantic labels named Graphite-23; 2) analyzes semantic segmentation architectures against the new data; and 3) contributes open source code for the community to progress research in degradation analysis of materials. A U-Net architecture with various backbones demonstrates competitive performance on the proposed dataset, with an mIoU up to 0.83, establishing a clear baseline for future research in this intersection of fields.

AB - When fuel materials for high-temperature gas-cooled nuclear reactors are quantification tested, significant analysis is required to establish their stability under various proposed accident scenarios, as well as to assess degradation over time. Typically, samples are examined by lab assistants trained to capture micrograph images used to analyze the degradation of a material. Analysis of these micrographs still require manual intervention which is time-consuming and can introduce human-error. While machine learning and computer vision models would be useful to this analysis, data for training such models is limited due to physical experiment costs, including lab hours and materials. This collaborative research are: 1) establishes an open dataset of micrographs and semantic labels named Graphite-23; 2) analyzes semantic segmentation architectures against the new data; and 3) contributes open source code for the community to progress research in degradation analysis of materials. A U-Net architecture with various backbones demonstrates competitive performance on the proposed dataset, with an mIoU up to 0.83, establishing a clear baseline for future research in this intersection of fields.

KW - Computer vision

KW - deep learning

KW - fuel analysis

KW - graphite

KW - neural networks

KW - nuclear materials

KW - semantic segmentation

UR - http://www.scopus.com/inward/record.url?scp=85174833970&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2023.3326074

DO - 10.1109/ACCESS.2023.3326074

M3 - Article

AN - SCOPUS:85174833970

SN - 2169-3536

VL - 11

SP - 118512

EP - 118520

JO - IEEE Access

JF - IEEE Access

ER -

Semantic Segmentation of Micrographs for Nuclear Fuel Analysis and Degradation Quantification

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this