In the three-dimensional biological microscopic image restoration processing, the selection of three dimensional point spread function (3D-PSF) space will determine the restoration effects and restoration time. Based on the theoretical analysis about the effect of 3D-PSF sampling interval and layers on biological image restoration effects and restoration time, through the simulation experiments, the relationship among 3D-PSF layers and biological image restoration effects and restoration time under the same 3D-PSF space size can be obtained. Experimental results showed that, for 3D-PSF with the same space size, the image restoration effect is improved with the increase of 3D-PSF layers, the degree of improvement is gradually slowed down with the increase of layers. Restoration time also increases with the increase of layers, and it shows an approximately linear increasing relationship. Based on the experimental results, Matlab fitting toolbox was used to construct the relationship model among 3D-PSF layers and normalized restoration time and Improved Signal-to-Noise Ratio (ISNR). The establishment of these models provided the theoretical basis and selection methods for 3D-PSF layer selection in the practical application of cell slice acquisition and three-dimensional biological microscopic image restoration.