Nuclear Sciences Division, Pacific Northwest National Laboratory, Richland, WA, 99352, USA.
Computer Science Department, Western Washington University, Bellingham, WA, 98225, USA.
Sci Rep. 2019 Sep 4;9(1):12744. doi: 10.1038/s41598-019-49105-0.
Crystalline materials exhibit long-range ordered lattice unit, within which resides nonperiodic structural features called defects. These crystallographic defects play a vital role in determining the physical and mechanical properties of a wide range of material systems. While computer vision has demonstrated success in recognizing feature patterns in images with well-defined contrast, automated identification of nanometer scale crystallographic defects in electron micrographs governed by complex contrast mechanisms is still a challenging task. Here, building upon an advanced defect imaging mode that offers high feature clarity, we introduce DefectSegNet - a new convolutional neural network (CNN) architecture that performs semantic segmentation of three common crystallographic defects in structural alloys: dislocation lines, precipitates and voids. Results from supervised training on a small set of high-quality defect images of steels show high pixel-wise accuracy across all three types of defects: 91.60 ± 1.77% on dislocations, 93.39 ± 1.00% on precipitates, and 98.85 ± 0.56% on voids. We discuss the sources of uncertainties in CNN prediction and the training data in terms of feature density, representation and homogeneity and their effects on deep learning performance. Further defect quantification using DefectSegNet prediction outperforms human expert average, presenting a promising new workflow for fast and statistically meaningful quantification of materials defects.
晶体材料表现出长程有序的晶格单元，其中存在称为缺陷的非周期性结构特征。这些晶体学缺陷在决定广泛的材料系统的物理和机械性能方面起着至关重要的作用。虽然计算机视觉已经成功地识别出具有明确定义对比度的图像中的特征模式，但在由复杂对比度机制控制的电子显微镜中自动识别纳米级晶体学缺陷仍然是一项具有挑战性的任务。在这里，我们在提供高特征清晰度的先进缺陷成像模式的基础上，引入了 DefectSegNet-一种新的卷积神经网络（CNN）架构，用于对结构合金中的三种常见晶体学缺陷进行语义分割：位错线、析出物和空隙。在一小部分高质量钢缺陷图像上进行监督训练的结果表明，所有三种类型的缺陷的像素精度都很高：位错为 91.60±1.77%，析出物为 93.39±1.00%，空隙为 98.85±0.56%。我们讨论了 CNN 预测和训练数据在特征密度、表示和均匀性方面的不确定性来源，以及它们对深度学习性能的影响。使用 DefectSegNet 预测进行进一步的缺陷量化优于人类专家的平均水平，为快速和具有统计学意义的材料缺陷量化提供了一种有前途的新工作流程。
