Microstructure-based classification and electronmicroscopy analysis of nanoporous metals by machine learning (B09)

通过机器学习对纳米多孔金属进行基于微观结构的分类和电子显微镜分析（B09）

• 批准号: 407600336
• 金额: --
• 依托单位: Institut für Kontinuums- und Werkstoffmechanik
• 依托单位国家: 德国
• 项目类别: Collaborative Research Centres
• 财政年份: 2018
• 资助国家: 德国
• 起止时间: 2017-12-31 至 2023-12-31
• 项目状态: 已结题
• 来源: https://gepris.dfg.de/gepris/projekt/407600336?language=en
• 关键词: Microstructure; based; classification; electronmicroscopy; analysis

Focused ion beam (FIB) tomography determines the three-dimensional microstructure of materials via a series of two-dimensional scanning electron microscope (SEM) images. In nanoporous metals, FIB tomography is often facing problems with so-called shine-through effects, which can significantly reduce the accuracy with which FIB tomography data can be segmented. This project will use machine learning to develop a new segmentation method which can efficiently suppress shine-through effects and thus enable a reconstruction even of complex hierarchical multi-scale microstructures of nanoporous metals with excellent accuracy.

聚焦离子束（FIB）层析成像通过一系列二维扫描电子显微镜（SEM）图像确定材料的三维微观结构。在纳米多孔金属中，FIB断层扫描通常面临所谓的穿透效应的问题，这会显著降低FIB断层扫描数据可以分割的准确性。该项目将使用机器学习开发一种新的分割方法，该方法可以有效地抑制穿透效应，从而能够以优异的精度重建纳米多孔金属的复杂层次多尺度微观结构。