Automated analysis of evolving interfaces during in situ electron microscopy

In situ electron microscopy allows one to monitor dynamical processes at high spatial and temporal resolution. This produces large quantities of data, and hence automated image processing algorithms are needed to extract useful quantitative measures of the observed phenomena. In this work, we outline an image processing workflow for the analysis of evolving interfaces imaged during liquid cell electron microscopy. As examples, we show metal electro-deposition at electrode surfaces; beam-induced nanocrystal formation and dissolution; and beam-induced bubble nucleation, growth, and migration. These experiments are used to demonstrate a fully automated workflow for the extraction of, among other things, interface position, roughness, lateral wavelength, local normal velocity, and the projected area of the evolving phase as functions of time. The relevant algorithms have been implemented in Mathematica and are available online.