High-quality reconstruction of capacitance tomography data is important to get the quantitative information from the cross sectional images of the multiphase pipe flow. Due to the complex nature of the capacitance sensors, the reconstruction algorithms well developed for medical tomography are not applicable. The main problems lie in two aspects. One is the ＇soft-field＇ effect, the other is the limited number of measurements. To resolve these problems and get high-quality images, a reconstruction algorithm, named simultaneous iterative reconstruction technique (SIRT), often used in geology investigation, is introduced. According to the fastness of the process tomography and the smoothing effect of SIRT, some improvements have been made. These methods are compared to the well-known linear back projection algorithm (LBP) and the linear back projection thresholded algorithm widely used in capacitance tomography. A computer simulated and real eight-electrode capacitance tomography system has been used for the evaluation. (C)2000 Elsevier Science S.A. All rights reserved.