The void fraction of gas–liquid two-phase flow has been visualized by fuzzy phase classification implemented in electrical impedance tomography (fEIT) . The fEIT consists of two steps which are image reconstruction step of from non-linear conductivity σ from measured voltage vexp and the fuzzy phase classification step of probabilistic cluster (gas or liquid) from reconstructed conductivity σ. The experiments carried out in a vertical pipe with an inner diameter D of 25 mm at liquid and gas superficial velocities Jl = 0, 0.2, 0.4 and 0.6 m/s for liquid, Jg = 0, 0.11, 0.32 and 0.53 m/s for gas. The fEIT provided a clear and distinct observation of flow patterns in the upstream region successfully. The accuracy of void fraction distribution α by the newly proposed fEIT is improved by 30% rather than the accuracy of void fraction distribution α by typical EIT.