Overview: Computer Vision and Machine Learning for Microstructural Characterization and Analysis

Abstract

Microstructural characterization and analysis is the foundation of microstructural science, connecting materials structure to composition, process history, and properties. Microstructural quantification traditionally involves a human deciding what to measure and then devising a method for doing so. However, recent advances in computer vision (CV) and machine learning (ML) offer new approaches for extracting information from microstructural images. This overview surveys CV methods for numerically encoding the visual information contained in a microstructural image using either feature-based representations or convolutional neural network (CNN) layers, which then provides input to supervised or unsupervised ML algorithms that find associations and trends in the high-dimensional image representation. CV/ML systems for microstructural characterization and analysis span the taxonomy of image analysis tasks, including image classification, semantic segmentation, object detection, and instance segmentation. These tools enable new approaches to microstructural analysis, including the development of new, rich visual metrics and the discovery of processing-microstructure-property relationships.

Development of Supplements Prevention System in Oxygen Converter Process in Order to Increase the Economic Efficiency of Steel Melting

“…Comparing the results of the analysis of smelting passports and the obtained mathematical models through the prism of the possibility of introducing them into a real production process with the results of foreign studies [6][7][8][9][10], we can say that the proposals developed in the article are at a high technological level.…”

“…In recent years, advanced methods have been developed to monitor and control processes in metallurgy. The introduction of powerful computing technology allowed technologists to accumulate and analyze large amounts of data in order to introduce advanced automation systems in metallurgy [6], and the development of programming languages allows introducing machine learning elements into metallurgy [7] to refine the developed models directly in the course of technological processes. Due to this, a lot of works have appeared devoted to the introduction into a real production process of models developed to control both the entire technology of steel production, and individual processes.…”

Efficiency and Accuracy of Simulated Microstructure Images Generated by Machine Learning

“…3). There are several widely apply and established algorithms used in machine learning methods [1,3,10]. The first group can include: weighted neighbourhood clustering that examples might be decision trees, random forest or k-Nearest neighbour, are applied.…”

“…Getting to know a material by discovering its structure and understanding the phenomena that allow it to shape its properties is essential for designing new materials [2,3].…”

“…Contemporary techniques of simulating material structures, using FEM methods, simulations using machine learning techniques or Markov processes, make it possible to obtain the desired material properties by designing their structure. The classic process of creating new materials has been reversed, which has opened up new possibilities for the development of modern materials and has broken the old design path [3,4].…”