Data-efficient U-Net for Segmentation of Carbide Microstructures in SEM Images of Steel Alloys
This enables rapid, automated carbide quantification for alloy design in metallurgy, with incremental improvements in data efficiency.
The paper tackled the problem of segmenting carbide microstructures in SEM images of steel alloys, which is difficult due to gray-value overlap, by developing a data-efficient U-Net trained on only 10 annotated images, achieving a Dice-Sørensen coefficient of 0.98 and significantly outperforming classical methods.
Understanding reactor-pressure-vessel steel microstructure is crucial for predicting mechanical properties, as carbide precipitates both strengthen the alloy and can initiate cracks. In scanning electron microscopy images, gray-value overlap between carbides and matrix makes simple thresholding ineffective. We present a data-efficient segmentation pipeline using a lightweight U-Net (30.7~M parameters) trained on just \textbf{10 annotated scanning electron microscopy images}. Despite limited data, our model achieves a \textbf{Dice-Sørensen coefficient of 0.98}, significantly outperforming the state-of-the-art in the field of metallurgy (classical image analysis: 0.85), while reducing annotation effort by one order of magnitude compared to the state-of-the-art data efficient segmentation model. This approach enables rapid, automated carbide quantification for alloy design and generalizes to other steel types, demonstrating the potential of data-efficient deep learning in reactor-pressure-vessel steel analysis.