Gamma-ray densitometry is widely implemented in oil industry because it is an online technique and also has a
good precision. If there is single phase flow in oil pipelines, measuring the density is possible just by using one
source and one detector. But if in addition to oil, there is gas in oil pipelines and in fact there is a two-phase flow,
conventional gamma ray densitometry (one source and one detector) could not be used for determining the
density of liquid phase. In this study, a novel method is proposed for online measuring density of liquid phase in
annular regime of liquid-gas two-phase flows using dual modality densitometry technique and artificial neural
network (ANN). An experimental setup was designed in order to provide the required input data for training and
testing the network. Registered counts in both scattering and transmission detectors were used as the inputs of
the ANN and density of liquid phase was used as the output of the ANN. Using the proposed methodology,
density of liquid phase was predicted with error of less than 0.031 g/cm−3 in annular regime of gas-liquid two
phase flows for void fractions in the range of 10–70 percentages.