Fast point-based hologram generation method using high-frequency information extraction

Computer-generated hologram based holographic 3D display is an important development direction of holographic 3D display. However, the heavy amount of calculation affects the generation speed of the holograms, which limits the further development of real-time holographic 3D display. In this paper, a fast point-based hologram generation method using high-frequency (HF) information extraction is proposed. Firstly, an image information processing method based on the Gaussian filter is proposed to separate the HF information and lowfrequency (LF) information of 3D object. Then, the HF hologram is calculated by using the novel-look-up-table algorithm, while the LF hologram is calculated by the angular spectrum algorithm. Finally, the final hologram is obtained by overlaying HF hologram and LF hologram. Compared with the traditional method, the proposed method improves the generation speed of the hologram by more than 65%. Moreover, the details of 3D object can be fully reconstructed, and the quality of the reconstructed image is slightly improved with lower speckle noise. Experimental results verify the feasibility of the proposed method.