Infrared and Visible Missile-borne Image Fusion Based on Structural Information

This paper proposed a structure based infrared visible light missile borne image fusion method. The method uses the idea of structural patch decomposition to decompose the image block into three components: signal strength, mean intensity and signal structure. The Laplace pyramid is used to decompose the signal structure component into low frequency band and high frequency band. Different methods are used to calculate the recovered signal structure. Finally, the fusion of image structure block decomposition is transformed into the final fusion image based on mean filtering. The fusion experiment is carried out by using the simulated missile-borne infrared and visible images, and the experimental results are evaluated subjectively and objectively. The experimental results show that the proposed method is superior to some classical fusion methods in subjective and objective assessment, and can obtain better fusion effect. The algorithm has high effectiveness and stability, and can be applied to real-time fusion tasks to a certain extent.