Role of length-scale in machine learning based image analysis of ductile fracture surfaces

Abstract

Recent advancements in machine learning (ML) techniques have opened up new opportu- nities for using image analysis to solve materials science problems. In this work, we have used an ML-based workflow to classify the fracture surfaces of dual-phase steels subjected to different stress states. This task is not straightforward, as the ductile fracture surfaces of many metallic materials exhibit similar features, such as dimples. The ML-based workflow uses a pre-trained convolution neural network in unsupervised mode to extract image fea- tures, which are then reduced in dimensionality using principal component analysis. Next, images are clustered and classified using K-Means and K-Nearest Neighbors algorithms, re- spectively. Our results show that the accuracy of the ML-based technique is sensitive to the length-scales of the fracture surface images, and the critical length-scale corresponding to the maximum accuracy depends on the typological categories being classified. A physical interpretation of the critical length-scales associated with the fracture surface images is pro- vided through quantitative fracture surface roughness analysis. Our work demonstrates the potential of using unsupervised ML-based techniques for fractography of ductile materials, especially for typological classification. More importantly, it emphasizes the importance of length-scales in image analysis in materials science.