In this paper,an infrared and visible image fusion method based on FGF(Fast Guided Filter)and VGGNet19 is proposed to address the problems of insufficiently prominent infrared targets and blurred edges in infrared and visible image fusion.Firstly,the source images are decomposed into two scales by fast guided filter.Secondly,the initial fusion weight map is obtained by using VGGNet19 network.Then,using OTSU(Maximum Interclass Variance)to extract the target and background of the initial fusion weight map,the binary image is obtained.Next,the fusion weights are obtained by using fast guided filter.Finally,the reconstruction is performed by addition,and the final fusion image is obtained by fast guided filter and image enhancement.The experimental results show that this method can an effectively retain significant thermal targets with clear edge details,and achieve better results in subjective as well as objective evaluation metrics.