An Infrared and Visible Image Fusion Algorithm Method Based on a Dual Bilateral Least Squares Hybrid Filter

Infrared and visible images of the same scene are fused to produce a image with richer information. However, most current image-fusion algorithms suffer from insufficient edge information retention, weak feature representation, poor contrast, halos, artifacts, can only be applied single scene. To address these issues, we propose novel infrared visual fusion algorithm based on bilateral–least-squares hybrid filter (DBLSF) least-squares bilateral model (BLF-LS). The proposed utilizes residual network ResNet50 adaptive strategy structure tensor fuse base detail layers decomposition, respectively. Experiments 32 sets TNO dataset show that, although our sacrifices overall time efficiency, Combination 1 approach better preserve integrity; reduce loss source features; suppress artifacts halos; compare favorably other in terms structural similarity, multiscale root mean square error, peak signal-to-noise ratio, correlation coefficient by at least 2.71%, 1.86%, 0.09%, 0.46%, 0.24%, 0.07%; 2 effectively improve contrast features enrich information, an average improvement 37.42%, 26.40%, 26.60% three metrics gradient, intensity, spatial frequency compared algorithms.