The ocean surface current (OSC) is one of the key factors which affect accuracy in OSC parameters inversed with a sequence of X-band radar images. Although OSC cannot be directly measured by instruments, this paper investigates the OSC inversion algorithm. In the process of OSC inversion, the energy cost function with constraint terms is established, thus resulting in OSC inversion formulae. The optimum solution can be derived by weighted least squares method. The inversion accuracy can be improved by taking into account the high-order dispersion relationship. In order to test the inversion algorithm, a sequence of radar images are acquired by X-band radar, both moving average and dispersion relationship are used to filter noises in the images. The algorithm is finally validated with the real X-band radar images, and the preliminary testing results show that the inversion approach is feasible.