Wavefront sensing from multiple focal plane images is a promising technique for high-contrast imaging systems. However, the wavefront error of an optics system can be properly reconstructed only when it is very small. This paper presents an iterative optimization algorithm for the direct measurement of large static wavefront errors from only one focal plane image. We first measure the intensity of the pupil image to get the pupil function of the system and acquire the aberrated image on the focal plane with a phase error that will be measured. Then we induce a dynamic phase on the tested pupil function and calculate the associated intensity of the reconstructed image on the focal plane. The algorithm will then try to minimize the intensity difference between the reconstructed image and the aberrated test image in the focal plane, where the induced phase is a variable of the optimization algorithm. The simulation shows that the wavefront of an optical system can theoretically be reconstructed with high precision, which indicates that such an iterative algorithm may be an effective way to perform wavefront sensing for high-contrast imaging systems.