Infrared and Visible Image Fusion based on Deep Gradient Constraint Driven Network

Ze Wang, Guoxia Xu

Abstract: Infrared and visible image fusion technology refers to the fusion of thermal radiation information and texture details contained in infrared image and visible image respectively, so as to obtain high-quality fused images, which has important applications in intelligent monitoring, target detection and tracking, etc. The two images are based on different imaging principles. How to fuse the advantages of their respective images and ensure that the images are not lost is the key of fusion technology. Different from the traditional algorithms that only focus on the global information of superimposed images, this paper proposes a fusion algorithm that pays more attention to the local gradient information of images to extract more detailed gradient information. Experiments show that our algorithm is more effective and advanced.