In order to preserve more edge and tex ture information of infrared and visible images,the guided filter is introduced to process high-frequency coefficients ased on the classical image fusion algorithm of non- subsampled shearlet transform (NSST) combined with Spiking Cortical Model,SCM),then it is used as the external excitation of SCM to fuse the high-frequency parts.Regional mean and regional energy are combined as the excitation of SCM,to fuse the low frequency part.Simulation results show that the fusion ef fect of the proposed algorithm is obviously better than other algorithms.In subjective visi on,infrared target is prominent,background information is rich,and edge and texture inform ation is preserved well.In objective evaluation,edge retention factor,information entr opy,contrast andother indicators areobviously improved.