U-Net++ based segmentation of textures across the CrB–CrE phase transition
- 30 September 2026
- journal article
- Published by Royal Society of Chemistry (RSC) in Soft Matter
- Vol. 22 (37) , 6264-6271
Abstract
Machine learning enables automated identification of liquid-crystalline phases from polarized optical microscopy textures, providing a rapid approach for tracking phase transitions.Keywords
Funding Information
- Narodowym Centrum Nauki (MINIATURA 8 project, grant no. 2024/08/X/ST3/00769)
This publication has 29 references indexed in Scilit:
- Deep Learning Analysis of Crystallization Using Polarized Light Microscopy and U-Net SegmentationThe Journal of Physical Chemistry B, 2025
- Learning phase transitions by siamese neural networkPhysica A: Statistical Mechanics and its Applications, 2025
- Key factors in neural network performance for liquid crystal texture classificationJournal of Molecular Liquids, 2025
- SCABNet : A Novel Polyp Segmentation Network With Spatial‐Gradient Attention and Channel PrioritizationInternational Journal of Imaging Systems and Technology, 2025
- A modified Adam algorithm for deep neural network optimizationNeural Computing and Applications, 2023
- A comparison of deep learning U-Net architectures for posterior segment OCT retinal layer segmentationScientific Reports, 2022
- Scientific Machine Learning Through Physics–Informed Neural Networks: Where we are and What’s NextJournal of Scientific Computing, 2022
- A systematic review on overfitting control in shallow and deep neural networksArtificial Intelligence Review, 2021
- Learning physical properties of liquid crystals with deep convolutional neural networksScientific Reports, 2020
- FT-IR and Two-Dimensional Correlation Analysis of the Liquid Crystalline Phase Transitions in the 4-Bromobenzylidene-4′-alkyloxyanilinesThe Journal of Physical Chemistry B, 2013