Infrared and visible image fusion using total variation model

Image fusion is a process of combining complementary information from multiple images of the same scene into an image, so that the resultant image contains a more accurate description of the scene than any of the individual source images. In this paper, we propose a novel fusion strategy for infrared (IR) and visible images based on total variation (TV) minimization. By constraining the fused image to have similar pixel intensities with the IR image and similar gradients with the visible image, it tends to simultaneously keep the thermal radiation and appearance information in the source images. We evaluate our method on a publicly available database with comparisons to other seven fusion methods. Our results have a major difference that the fused images look like sharpened IR images with detailed appearance information. The quantitative results demonstrate that our method also can achieve comparable metric values with other state-of-the-art methods.