The work aims to screen the hologram to realize the binarization of the computational hologram and apply the computational hologram to the printing field. An overall scheme of frequency modulation screening for computational hologram was designed and the effects of different screening algorithms on the quality of holographic reproduction image were discussed. Firstly, three types of grayscale images were holographically encoded to obtain holograms and then these holograms were subject to frequency modulation screening to gain binarized holograms by the error diffusion and dithering algorithms. Finally, the binarized holograms were transformed holographic reproduction images by means of light field. According to the comparison between the peak signal-to-noise ratio (PSNR) and the structure similarity index measure (SSIM) data of holographic reproduction images, the error diffusion algorithm was more suitable for computing holographic binarization and the dithering screening caused periodic patterns on the holograms, which resulted in frequency mixing phenomena on the holographic reproduction images, so the quality of the holographic reproduction images was reduced. Screening degrades the quality of holographic reproduction images. The error diffusion algorithm has better reproduction images and is suitable for hologram binarization. Furthermore, the dithering algorithm produces frequency mixing, so it is not suitable for hologram binarization.