Image optimization and phase recovery of metasurface holography based on improved Gerchberg-Saxton algorithm

In this study, a phase recovery method based on an improved Gerchberg-Saxton (GS) algorithm is proposed to optimize the design and reconstruction of metasurface holography. Firstly, CST electromagnetic simulation software is used to simulate the metasurface basic structure unit, obtain its complex amplitude transmission matrix, and construct 100×100 metasurface hologram by polynomial fitting method. In the phase recovery process, random phase perturbation and weighted intensity matching strategy are introduced to improve the convergence and robustness of the algorithm. The experimental results show that the proposed method can effectively improve the holographic image quality, optimize the phase recovery process, and enhance the manufacturing feasibility of metasurface optics, which provides important support for the design of high-precision metasurface holographic imaging.