TY - JOUR
T1 - Automatic detection of grains in partially recrystallized microstructures using deep learning
AU - Lin, Fengxiang
AU - Fang, Haixing
AU - Liu, Hong
AU - Zhang, Yubin
AU - Juul Jensen, Dorte
AU - Hovad, Emil
N1 - Publisher Copyright:
© 2024
PY - 2025
Y1 - 2025
N2 - Precise identification of recrystallizing grains in partially recrystallized microstructures is essential to obtain quantitative information regarding the recrystallization process. Automatic, robust, user-friendly, and unbiased identification methods that do not rely on hard-coded, preselected values would be highly advantageous. In this study, we test convolutional neural network instance segmentation models to achieve automatic segmentation of individual recrystallizing grains in partially recrystallized microstructures. Our training dataset includes micrographs obtained using electron backscattered diffraction from five alloys with different thermal-mechanical histories and more than 100,000 recrystallizing grains. We adapt and train two state of the art deep learning models, namely Mask R-CNN and PointRend. Both models provide instance segmentation results of good quality, enabling quantitative determination of the microstructural parameters. The PointRend model demonstrates better performance for grains with irregular shapes than Mask R-CNN. Compared to conventional methods, the trained deep learning approach is easier to use, more flexible, and applicable to a wide range of materials.
AB - Precise identification of recrystallizing grains in partially recrystallized microstructures is essential to obtain quantitative information regarding the recrystallization process. Automatic, robust, user-friendly, and unbiased identification methods that do not rely on hard-coded, preselected values would be highly advantageous. In this study, we test convolutional neural network instance segmentation models to achieve automatic segmentation of individual recrystallizing grains in partially recrystallized microstructures. Our training dataset includes micrographs obtained using electron backscattered diffraction from five alloys with different thermal-mechanical histories and more than 100,000 recrystallizing grains. We adapt and train two state of the art deep learning models, namely Mask R-CNN and PointRend. Both models provide instance segmentation results of good quality, enabling quantitative determination of the microstructural parameters. The PointRend model demonstrates better performance for grains with irregular shapes than Mask R-CNN. Compared to conventional methods, the trained deep learning approach is easier to use, more flexible, and applicable to a wide range of materials.
KW - Computer vision
KW - Convolutional neural network
KW - Deep learning
KW - EBSD
KW - Instance segmentation
KW - Microstructure
KW - Recrystallization
U2 - 10.1016/j.matchar.2024.114576
DO - 10.1016/j.matchar.2024.114576
M3 - Journal article
AN - SCOPUS:85210621705
SN - 1044-5803
VL - 219
JO - Materials Characterization
JF - Materials Characterization
M1 - 114576
ER -