In this repository, a Deep-learning model is given, which was trained to extract the vertical position of the gas-liquid-interface over time (liquid level time series) from video recordings of horizontal gas-liquid pipe flow experiments. Also called **L**iquid-gas **I**nterface **S**election **A**lgorithm (**LISA**). The xperiments were conducted in the european research projects MultiFlowMet I and II [see [MFM I][1] and [MFM II][2]].
In this repository, a Deep-learning model is given, which was trained to extract the vertical position of the gas-liquid-interface over time (liquid level time series) from video recordings of horizontal gas-liquid pipe flow experiments. Also called **L**iquid-gas **I**nterface **S**election **A**lgorithm (**LISA**). The Experiments were conducted in the european research projects MultiFlowMet I and II [see [MFM I][1] and [MFM II][2]].