TU Darmstadt / ULB / TUprints

A deep learned nanowire segmentation model using synthetic data augmentation

Lin, Binbin ; Emami, Nima ; Santos, David A. ; Luo, Yuting ; Banerjee, Sarbajit ; Xu, Bai-Xiang (2022):
A deep learned nanowire segmentation model using synthetic data augmentation. (Publisher's Version)
In: npj Computational Materials, 8, Springer, e-ISSN 2057-3960,
DOI: 10.26083/tuprints-00021425,
[Article]

[img] Text
s41524-022-00767-x.pdf
Copyright Information: CC BY 4.0 International - Creative Commons, Attribution.

Download (3MB)
[img] Text
21425Suppl.pdf
Copyright Information: CC BY 4.0 International - Creative Commons, Attribution.

Download (935kB)
Item Type: Article
Origin: Secondary publication via sponsored Golden Open Access
Status: Publisher's Version
Title: A deep learned nanowire segmentation model using synthetic data augmentation
Language: English
Abstract:

Automated particle segmentation and feature analysis of experimental image data are indispensable for data-driven material science. Deep learning-based image segmentation algorithms are promising techniques to achieve this goal but are challenging to use due to the acquisition of a large number of training images. In the present work, synthetic images are applied, resembling the experimental images in terms of geometrical and visual features, to train the state-of-art Mask region-based convolutional neural networks to segment vanadium pentoxide nanowires, a cathode material within optical density-based images acquired using spectromicroscopy. The results demonstrate the instance segmentation power in real optical intensity-based spectromicroscopy images of complex nanowires in overlapped networks and provide reliable statistical information. The model can further be used to segment nanowires in scanning electron microscopy images, which are fundamentally different from the training dataset known to the model. The proposed methodology can be extended to any optical intensity-based images of variable particle morphology, material class, and beyond.

Journal or Publication Title: npj Computational Materials
Volume of the journal: 8
Place of Publication: Darmstadt
Publisher: Springer
Collation: 12 Seiten
Classification DDC: 600 Technik, Medizin, angewandte Wissenschaften > 620 Ingenieurwissenschaften
Divisions: 11 Department of Materials and Earth Sciences > Material Science
11 Department of Materials and Earth Sciences > Material Science > Mechanics of functional Materials
Date Deposited: 07 Jun 2022 12:14
Last Modified: 22 Aug 2022 08:29
DOI: 10.26083/tuprints-00021425
Corresponding Links:
URN: urn:nbn:de:tuda-tuprints-214250
Additional Information:

Keywords: Characterization and analytical techniques, imaging techniques, optical spectroscopy

URI: https://tuprints.ulb.tu-darmstadt.de/id/eprint/21425
PPN: 495423548
Export:
Actions (login required)
View Item View Item