What question did this study set out to answer?

This study aims to enhance the semantic segmentation of microstructural phases in steel specimens using CNN architectures.

July 3, 2026Open Access

Benchmarking Convolutional Neural Network Architectures for Multi-Phase Semantic Segmentation: Challenges in Resolving Widmanstätten Ferrite Within Ferritic–Pearlitic Matrices

Key Points

This study aims to enhance the semantic segmentation of microstructural phases in steel specimens using CNN architectures.
Evaluated five CNN architectures: ResNet-152, Xception, SE-ResNet-50, DenseNet-169, EfficientNet-B5.
Utilized a dataset of 20 light optical microscopy images (LOMs) with manually annotated masks.
Assessed segmentation performance using three weight initialization strategies: ImageNet, MicroNet, and combined.
DenseNet-169 pretrained on ImageNet achieved the best patch-level performance for Widmanstätten ferrite segmentation (Dice: 50.75% ± 3.38%, IoU: 38.45% ± 3.01%).
Xception pretrained on ImageNet, after inference-based aggregation, showed improved metrics (Dice: 68.74% ± 1.12%, IoU: 52.52% ± 1.28%).
MAE for Xception performance was recorded at 4.55% ± 0.80%.

Abstract

The welded bead bending test (WBBT) serves as a pivotal procedure for evaluating the crack-arrest capacity of structural steels used for construction of safety-critical infrastructure according to ZTV-ING Part 4 or Deutsche Bahn Standard 918 002-02. Previous research has established that light optical microscopy images (LOMs) of specimens that do not pass the WBBT are frequently characterised by a high prevalence of Widmanstätten ferrite within the base material. While convolutional neural networks (CNNs) have successfully classified WBBT outcomes based on LOMs, the implemented approaches did not enable simultaneous pixel-wise delineation required for accurate quantification of Widmanstätten ferrite. However, the tonal similarity between Widmanstätten ferrite and polygonal ferrite renders conventional intensity-based thresholding ineffective for differentiation, necessitating a morphology-based segmentation approach. The present study proposes an automated semantic segmentation framework for the precise delineation of three microstructural phases present within the WBBT LOMs: polygonal ferrite, pearlite, and Widmanstätten ferrite. Utilising a dataset of 20 LOMs and corresponding manually annotated masks, five UNet encoder backbones were evaluated: ResNet-152, Xception, SE-ResNet-50, DenseNet-169, and EfficientNet-B5. To identify the optimal configuration for this microstructural segmentation task, each architecture was assessed using three distinct weight initialisation strategies: (A) ImageNet, (B) MicroNet, and (C) a combined approach. For Widmanstätten ferrite segmentation at patch level, DenseNet-169 pretrained on ImageNet achieves the best performance (Dice: 50.75% ± 3.38%, IoU: 38.45% ± 3.01%). Following inference-based aggregation, Xception pretrained on ImageNet yields improved results (Dice: 68.74% ± 1.12%, IoU: 52.52% ± 1.28%), with an MAE of 4.55% ± 0.80%.

Read Full Paperexternally

Demander à l'IA

Bookmark

View Full Paper

Cite This Study

Backofen et al. (Wed,) studied this question.

synapsesocial.com/papers/6a4752775c29257aa2579359 https://doi.org/https://doi.org/10.3390/met16070726

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

Demander à l'IA

Bookmark

View Full Paper