EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep-learning-based pyramid-transformer for localized porosity analysis of hot-press sintered ceramic paste

Zhongyi Xia iD, Boqi Wu iD, C Y Chan, Tianzhao Wu, Man Zhou, Ling Bing Kong iD

PLOS ONE · 2024

DOI: 10.1371/journal.pone.0306385 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS ONE, 2024. Verfasst von Zhongyi Xia und 5 weiteren Autor:innen. Beteiligte Einrichtung: College of Applied Technology, Shenzhen University, Shenzhen, Guangdong, China. Hauptthemen laut MEDLINE: Ceramics, Deep Learning, Microscopy, Electron, Scanning. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Jilin Provincial Department of Science and Technology Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Scanning Electron Microscope (SEM) is a crucial tool for studying microstructures of ceramic materials. However, the current practice heavily relies on manual efforts to extract porosity from SEM images. To address this issue, we propose PSTNet (Pyramid Segmentation Transformer Net) for grain and pore segmentation in SEM images, which merges multi-scale feature maps through operations like recombination and upsampling to predict and generate segmentation maps. These maps are used to predict the corresponding porosity at ceramic grain boundaries. To increase segmentation accuracy and minimize loss, we employ several strategies. (1) We train the micro-pore detection and segmentation model using publicly available Al 2 O 3 and custom Y 2 O 3 ceramic SEM images. We calculate the pixel percentage of segmented pores in SEM images to determine the surface porosity at the corresponding locations. (2) Utilizing high-temperature hot pressing sintering, we prepared and captured scanning electron microscope images of Y 2 O 3 ceramics, with which a Y 2 O 3 ceramic dataset was constructed through preprocessing and annotation. (3) We employed segmentation penalty cross-entropy loss, smooth L1 loss, and structural similarity (SSIM) loss as the constituent terms of a joint loss function. The segmentation penalty cross-entropy loss helps suppress segmentation loss bias, smooth L1 loss is utilized to reduce noise in images, and incorporating structural similarity into the loss function computation guides the model to better learn structural features of images, significantly improving the accuracy and robustness of semantic segmentation. (4) In the decoder stage, we utilized an improved version of the multi-head attention mechanism (MHA) for feature fusion, leading to a significant enhancement in model performance. Our model training is based on publicly available laser-sintered Al 2 O 3 ceramic datasets and self-made high-temperature hot-pressed sintered Y 2 O 3 ceramic datasets, and validation has been completed. Our Pix Acc score improves over the baseline by 12.2%, 86.52 vs. 76.01, and the mIoU score improves from by 25.5%, 69.10 vs. 51.49. The average relative errors on datasets Y 2 O 3 and Al 2 O 3 were 6.9% and 6.36%, respectively.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

CeramicsDeep LearningMicroscopy, Electron, ScanningPorosityHot TemperatureAluminum Oxide

Substanzen

Aluminum Oxide

Autor:innen und Institutionen

  • College of Applied Technology, Shenzhen University, Shenzhen, Guangdong, China.
  • College of New Materials and New Energies, Shenzhen Technology University, Shenzhen, Guangdong, China.
  • Key Laboratory for Comprehemsive Energy Saving of Cold Regions Architecture of Ministry of Education, Jilin Jianzhu University, Changchun, Jilin, China.

ORCID-Profile: Zhongyi Xia, Boqi Wu, Ling Bing Kong

Förderung

  • Jilin Provincial Department of Science and Technology Project

Zitieren

APA

Xia, Z., Wu, B., Chan, C. Y., Wu, T., Zhou, M., & Kong, L. B. (2024). Deep-learning-based pyramid-transformer for localized porosity analysis of hot-press sintered ceramic paste. PLOS ONE. https://doi.org/10.1371/journal.pone.0306385

Vancouver

Xia Z, Wu B, Chan CY, Wu T, Zhou M, Kong LB. Deep-learning-based pyramid-transformer for localized porosity analysis of hot-press sintered ceramic paste. PLOS ONE. 2024. doi:10.1371/journal.pone.0306385

BibTeX
@article{Xia2024,
  author = {Xia, Zhongyi and Wu, Boqi and Chan, C Y and Wu, Tianzhao and Zhou, Man and Kong, Ling Bing},
  title = {Deep-learning-based pyramid-transformer for localized porosity analysis of hot-press sintered ceramic paste},
  journal = {PLOS ONE},
  year = {2024},
  doi = {10.1371/journal.pone.0306385}
}
Kostenlos starten →