Infrared and visible image fusion for shipborne electro-optical pod in maritime environment

In recent years, electro-optical pod (EOP) including infrared and visible sensors has been widely applied in the surveillance system of maritime surveillance ships. The infrared and visible image fusion is one of the efficient ways to help surveillance systems to make next-step decisions. However, unlike normal surveillance tasks, maritime infrared and visible image fusion is a quite challenging because of strong illumination and complex weather. In this paper, we propose a novel infrared and visible image fusion method for shipborne EOP in maritime environment. Specifically, we first design a gray scale projection model to construct the background according to the infrared imaging link of sea scene. In addition, a regional intensity difference strategy is developed for obtaining infrared targets, and then the target regions are added to the background to reconstruct the infrared degradation image. Finally, since the maritime visible image lacks texture and target information, we study a color transfer mechanism based on the subspace domain, which transfers the color features of the visible image to the infrared degraded image for obtaining the fused image. Experiments in various maritime environments show that the fused image of the proposed fusion method not only preserves important search targets, but also has the look and feel of visible images. Not only that, the proposed fusion method also has better performance than seven state-of-the-art comparison fusion method.