Extracting ductile cast iron microstructure parameters from fracture surfaces: A deep learning based instance segmentation approach

Engineering Fracture Mechanics, vol. 311, pp. 110586

Abstract

• Deep-learning based microstructural analysis of ductile cast iron fracture surfaces. • Mask R-CNN model trained for graphite particle detection using SEM images. • Fracture surface topography restored using 4-quadrant BSE signal. • Key parameters as particle size, shape, and distance extracted accurately. • Subsequent probabilistic simulation of crack resistance curves. This study investigates the deep-learning based microstructural analysis from SEM images of ductile cast iron fracture surfaces. A Mask R-CNN model was trained, achieving 70% precision and 75% recall in graphite particle detection. Combined with a fracture surface reconstruction using the. 4-quadrant backscattered electron signal, key parameters, including the particle size, shape and distance were extracted accurately. Compared to micrograph analysis, following probabilistic simulations showed the impact of the higher microstructural variance for the fracture surfaces on crack initiation, leading to higher scatter and elevated crack resistance curves. This highlights the potential of deep-learning based analysis for comprehensive microstructural characterization.

Authors 4

1. Johannes Tlatlik

   Fraunhofer Institute for Mechanics of Materials

   Affiliation as printed

   Fraunhofer Institute for Mechanics of Materials, Wöhlerstraße 11, D-79108 Freiburg, Germany

2. Carla Beckmann

   Fraunhofer Institute for Mechanics of Materials

   Affiliation as printed

   Fraunhofer Institute for Mechanics of Materials, Wöhlerstraße 11, D-79108 Freiburg, Germany

3. Benedikt Rohrmüller

   Fraunhofer Institute for Mechanics of Materials

   Affiliation as printed

   Fraunhofer Institute for Mechanics of Materials, Wöhlerstraße 11, D-79108 Freiburg, Germany

4. Sebastian Münstermann Aachen Institute of Metal Forming

   RWTH Aachen University

   Affiliation as printed

   Institute of Metal Forming, RWTH Aachen University, Intzestraße 10, D-52072 Aachen, Germany