From Microstructure to Performance Optimization: Innovative Applications of Computer Vision in Materials Science

The rapid advancement of computer vision (CV) technology is transforming traditional approaches to material microstructure analysis. This review outlines the history of CV and explores the applications of deep learning (DL)-driven CV in four key areas of materials science: microstructure-based performance prediction, microstructure information generation, microstructure defect detection, and crystal structure-based property prediction. CV has significantly reduced the cost of traditional experimental methods in material performance prediction. Moreover, recent progress in CV for generating microstructure images and detecting defects has led to more efficient and reliable assessments of material performance. By integrating predictions based on both crystal and microstructure data, DL-driven CV models are accelerating the design of new materials with optimized performance, thereby paving the way for the discovery and innovation of next-generation materials. Finally, insights and future prospects are provided regarding the rapid interdisciplinary development in this field.