Segmentation of backscattered electron images of geopolymers using convolutional autoencoder network

Shohreh Sheiati*, Sanaz Behboodi, Navid Ranjbar*

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

52 Downloads (Pure)


Segmentation of backscattered electron (BSE) images of cementitious materials is often used to quantify different microstructural features for the sake of performance estimation at macro-scale levels. However, the heterogeneous nature of cementitious matrices compounds with varying imaging conditions can lead the conventional segmentation methods to a processing bottleneck for largescale experiments. To overcome these challenges, in this study, we evaluate the potential of deep autoencoder convolutional networks, specifically SegNet, for automatic segmentation of fly ash-based geopolymer images. We present the SegNet power in achieving a comparable accuracy to the human performance even with a few BSE images in the model’s training. The SegNet demonstrates magnification independent training that enables test image processing with both seen and unseen magnification levels. A comparative study shows that SegNet outperforms the Gaussian method on uncontrolled imaging conditions such as background brightness levels. In addition, we demonstrate the self-learning capability of SegNet in poorly annotated areas.
Original languageEnglish
Article number117846
JournalExpert Systems with Applications
Number of pages11
Publication statusPublished - 2022


  • Geopolymer
  • Backscattered electron imaging
  • Segmentation
  • Deep learning
  • Convolutional neural networks


Dive into the research topics of 'Segmentation of backscattered electron images of geopolymers using convolutional autoencoder network'. Together they form a unique fingerprint.

Cite this