We demonstrate generalizable semantic segmentation using minimal ground truth data.Correlated scanning electron microscopy(SEM)images and electron backscatter diffraction(EBSD)measurements of friction-stir processed 3...We demonstrate generalizable semantic segmentation using minimal ground truth data.Correlated scanning electron microscopy(SEM)images and electron backscatter diffraction(EBSD)measurements of friction-stir processed 316L stainless steel plates were used to train deep learning models for grain boundary segmentation.Secondary electron(SE)imaging taken at 10 keV correlated to EBSD-derived grain boundaries produced the best performing model.Notably,an ensemble of three models trained on a single SE image produced accurate segmentation over a series of backscatter electron(BSE)images of samples manufactured under different processing parameters,with a mean absolute error in grain size of 0.34μm.The generalizability of the models likely results from the similar escape depths of the SE training input and the EBSD training output and the reduced probability of stored strain artifacts appearing in the image.This highlights the importance of considering the physical principles behind imaging to develop robust models for microstructure characterization.展开更多
文摘We demonstrate generalizable semantic segmentation using minimal ground truth data.Correlated scanning electron microscopy(SEM)images and electron backscatter diffraction(EBSD)measurements of friction-stir processed 316L stainless steel plates were used to train deep learning models for grain boundary segmentation.Secondary electron(SE)imaging taken at 10 keV correlated to EBSD-derived grain boundaries produced the best performing model.Notably,an ensemble of three models trained on a single SE image produced accurate segmentation over a series of backscatter electron(BSE)images of samples manufactured under different processing parameters,with a mean absolute error in grain size of 0.34μm.The generalizability of the models likely results from the similar escape depths of the SE training input and the EBSD training output and the reduced probability of stored strain artifacts appearing in the image.This highlights the importance of considering the physical principles behind imaging to develop robust models for microstructure characterization.
