Microstructure segmentation with deep learning encoders pre-trained on a large microscopy dataset  

作　　者：Joshua Stuckner Bryan Harder Timothy M.Smith 

机构地区：[1]Materials and Structures Division,NASA Glenn Research Center,21000 Brookpark Rd,Cleveland,OH,44135,USA

出　　处：《npj Computational Materials》2022年第1期1905-1916,共12页计算材料学（英文）

基　　金：This work was supported by the NASA Transformational Tools and Technologies(TTT)project under the Transformative Aeronautics Concept Program within the Aeronautics Research Mission Directorate.

摘　　要：This study examined the improvement of microscopy segmentation intersection over union accuracy by transfer learning from a large dataset of microscopy images called MicroNet.Many neural network encoder architectures were trained on over 100,000 labeled microscopy images from 54 material classes.These pre-trained encoders were then embedded into multiple segmentation architectures including UNet and DeepLabV3+to evaluate segmentation performance on created benchmark microscopy datasets.Compared to ImageNet pre-training,models pre-trained on MicroNet generalized better to out-of-distribution micrographs taken under different imaging and sample conditions and were more accurate with less training data.When training with only a single Ni-superalloy image,pre-training on MicroNet produced a 72.2%reduction in relative intersection over union error.These results suggest that transfer learning from large in-domain datasets generate models with learned feature representations that are more useful for downstream tasks and will likely improve any microscopy image analysis technique that can leverage pre-trained encoders.

关 键 词：UNION INTERSECTION image 

分 类 号：TN762[电子电信—电路与系统]