The surface charge measurement method based on Pockels effect has been used to investigate the surface discharge and plasma phenomena due to its remarkable advantage. However, it remains an important subject to obtain the accurate results of surface charge distribution. In this paper the inversion algorithm research on charge density calculation in surface charge measurement based on Pockels effect was carried out. Two-dimensional Fourier transform and Wiener filter techniques were employed in the algorithm and the relationship between potential measured by Pockels crystal and surface charge density was processed in spatial frequency domain. This improved method considers the influence of sample's thickness and permittivity, and it also considers the impact of charges at different positions. Therefore, it is a better solution in calculating the charge density on different samples with different thickness. Besides, Wiener filter plays an important role in this algorithm, which reduces the system noise and improves the stability of numerical calculation. © 2011 Chinese Society for Electrical Engineering.