@article{e233ad1e47a14afc81535b3eead38d28,
title = "Optimizing convolutional neural networks to perform semantic segmentation on large materials imaging datasets: X-ray tomography and serial sectioning",
abstract = "Machine learning was used to segment large materials science datasets resulting from synchrotron-based x-ray computed tomography (XCT) images of dendrite growth, and serial sectioning (SS) images of dendrite coarsening. Both neural networks (NNs) yielded quantitatively more accurate outputs than conventional segmentation techniques using only 30 XCT or 6 SS training images. We show that performance can be improved if NNs are trained using a large number of small images that are sampled from the fixed amount of training data. The optimal image size and number of training images was identified for the XCT and SS datasets. NN transferability was also tested by applying the highest performing XCT and SS NNs to related datasets. While the initial segmentations were successful, applying simple transformations to the raw images further improved NN performance. These results show the great predictive ability and promising future of using machine learning for segmentation of large materials science datasets.",
keywords = "Convolutional neural network, Machine learning, Semantic segmentation, Serial sectioning, Solidification, X-ray tomography",
author = "Tiberiu Stan and Thompson, {Zachary T.} and Voorhees, {Peter W.}",
note = "Funding Information: The authors thank M. Rappaz (EPFL) and P. Jarry (Constellium) for fabricating the Al-Zn alloy, X. Xiao (BNL) for aid in XCT data acquisition, and E. Holm (CMU) for fruitful discussions. The XCT portion of this work was supported by the U.S. Department of Energy's Office of Science under grant number DE-FG02-99ER45782, and by the U.S. Department of Commerce, National Institute of Standards and Technology as part of the Center for Hierarchical Materials Design (CHiMaD) under the financial assistance award 70NANB14H012. The SS portion of this work was sponsored by National Aeronautics and Space Administration under grant number NNX16AR13G. The raw data (XCT and SS images) required to reproduce these findings are available to download from [http://doi.org/10.18126/M2RM08]. The processed data (ground truth segmentations) required to reproduce these findings are available to download from [http://doi.org/10.18126/M2W93J]. Funding Information: The authors thank M. Rappaz (EPFL) and P. Jarry (Constellium) for fabricating the Al-Zn alloy, X. Xiao (BNL) for aid in XCT data acquisition, and E. Holm (CMU) for fruitful discussions. The XCT portion of this work was supported by the U.S. Department of Energy 's Office of Science under grant number DE-FG02-99ER45782 , and by the U.S. Department of Commerce, National Institute of Standards and Technology as part of the Center for Hierarchical Materials Design (CHiMaD) under the financial assistance award 70NANB14H012 . The SS portion of this work was sponsored by National Aeronautics and Space Administration under grant number NNX16AR13G . Publisher Copyright: {\textcopyright} 2020 Elsevier Inc.",
year = "2020",
month = feb,
doi = "10.1016/j.matchar.2020.110119",
language = "English (US)",
volume = "160",
journal = "Materials Characterization",
issn = "1044-5803",
publisher = "Elsevier Inc.",
}