Multiphase distributions are common in industrial processes, such as the gas/oil/water three-phase flow, where the visualization of three media is of great significance for state monitoring and fault prediction of industrial production processes. Industrial process tomography possesses unique advantages of non-invasive, non-radiation, and low-cost in multiphase distribution imaging, attracting more and more attention. Due to the complexity of the physical parameters of gas/oil/water three media, traditional single-modality tomography technology is usually difficult to reconstruct and distinguish three media simultaneously. To overcome this problem, an electrical/ultrasonic dual-modality tomography method is proposed, where a coplanar dual-modality sensor is designed to obtain electrical and ultrasonic measurement data. To achieve image fusion, the same mesh model is used in electrical resistance tomography (ERT) and ultrasonic transmission tomography (UTT) inverse problems. The logistic regression and two-dimensional wavelet-transform fusion methods are proposed to fuse the ERT and UTT reconstructed images and visualize the gas/oil/water three media. Numerical and experimental tests are conducted, and image reconstruction results show that the proposed coplanar electrical/ultrasonic dual-modality tomography method can simultaneously reconstruct and distinguish three media in water continuous gas/oil/water three-phase distribution. Especially, the proposed logistic regression and two-dimensional wavelet-transform fusion methods can also improve the imaging quality compared with the traditional single-modality and weighted-average fusion methods.