In order to solve the problem of missing object information due to external influence in the sampling process of computational ghost imaging, wavelet analysis is applied to the computational ghost imaging system, and image fusion method is introduced into this system to build a computational ghost imaging system based on image fusion. Simulation and experimental research are carried out on this problem. The results show that this system can effectively recover the information of the object to be measured and improve the imaging quality, which is more suitable for the actual application environment than the traditional computational ghost imaging system.