Simple automation of SEM-EDS spectral maps analysis with Python and the edxia framework

Fabien Georget; William Wilson; Karen L Scrivener

doi:10.1111/jmi.13099

Simple automation of SEM-EDS spectral maps analysis with Python and the edxia framework

J Microsc. 2022 May;286(2):185-190. doi: 10.1111/jmi.13099. Epub 2022 Mar 24.

Authors

Fabien Georget^{1

2}, William Wilson^{1

3}, Karen L Scrivener¹

Affiliations

¹ Laboratory of Construction Materials, EPFL, Lausanne, Switzerland.
² Institute of Building Materials, RWTH Aachen, Aachen, Germany.
³ Université de Sherbrooke, Sherbrooke, Quebec, Canada.

Abstract

In a recent article, we described the edxia framework, a user-friendly framework to analyse the microstructure of cementitious materials using SEM-EDS hypermaps. The manual approach presented was shown to be efficient to answer the relevant scientific questions. However, it is limited for batch analysis and (semi-)automated treatments. In this article, we show how the framework can be used to customise the analysis to the problem at hand. We first present some possible extensions, and then we provide a simple example of automatic clustering, using the flexible Python scientific libraries which will allow to define more custom workflows in the future.

Keywords: SEM-EDX; automation; cement paste; edxia; image analysis.