Image regression analysis for linking the microstructure and property of steel

Abstract

An image regression analysis using a convolutional neural network (CNN) was conducted to correlate microstructural images with the mechanical properties of tempered martensitic steel. The proposed model could link the microstructure with a property without any manual feature extraction. To demonstrate the mechanism of the CNN−based image regression algorithm, its convoluted output was dimensionally reduced using uniform manifold approximation and projection (UMAP) and then plotted.

Keywords

• Image regression
• Convolutional neural network
• UMAP
• Microstructure–property relationship
• Explainable AI
• Steel