Share Email Print
cover

Proceedings Paper • new

Building towards a universal neural network to segment large materials science imaging datasets
Author(s): Tiberiu Stan; Zachary T. Thompson; Peter W. Voorhees
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Segmentation of large images can be one of the most time-consuming steps in the analysis of materials science datasets. Convolutional neural networks (NNs) have been shown to reduce segmentation time compared to manual techniques, but training a new NN is often required for each dataset. We show that simply combining NN training datasets does not necessarily lead to a NN capable of segmenting multiple types of images. In the present study, we first show that SegNet-based neural networks (NNs) can be trained to accurately segment Al-Zn x-ray computed tomography and Pb-Sn serial sectioning images. Applying the Al-Zn NN to the Pb-Sn test image led to misclassified smudges as dendrites, and misclassified speckles as background. Applying the Pb-Sn NN to the Al-Zn test image was unsuccessful, likely because the Al-Zn dendrites had a higher luminance than the Pb-Sn dendrites. The Mix NN (trained using the combined Al-Zn and Pb-Sn datasets) was better at segmenting the Pb-Sn test image than the Al-Zn test image. This is likely because the Pb-Sn training dataset contained ~4.5 times as many dendrite pixels as the Al-Zn training dataset, thus the Mix NN was over-tuned to identify Pb-Sn dendrites. Simply combining the training datasets was overall detrimental to NN performance, but assigning different classes to the Al-Zn and Pb-Sn dendrites may lead to enhanced performance in the future. These findings serve as guidelines in the quest to develop a universal NN for segmentation of large materials science datasets.

Paper Details

Date Published: 10 September 2019
PDF: 6 pages
Proc. SPIE 11113, Developments in X-Ray Tomography XII, 111131G (10 September 2019); doi: 10.1117/12.2525290
Show Author Affiliations
Tiberiu Stan, Northwestern Univ. (United States)
Zachary T. Thompson, Northwestern Univ. (United States)
Peter W. Voorhees, Northwestern Univ. (United States)


Published in SPIE Proceedings Vol. 11113:
Developments in X-Ray Tomography XII
Bert Müller; Ge Wang, Editor(s)

© SPIE. Terms of Use
Back to Top