Visible And Infrared Image Fusion Algorithm

Image fusion technique is an image processing technique which put the image data of the same scene in different spectrum or focuses into one image. The fused image can make a better visibility and contains complementary information of source images. This technique can integrate the advantage of multiple imaging devices. Visible and infrared image fusion technique was often used, for visible image is rich in detail, at the same time, infrared image can provide a clear outline of object and also can find hidden object. That means they have complementary information. Visible and infrared image fusion technique has a wide range of applications, and also has great value at battlefield auxiliary at night and car navigation.The study in this article is based on visible and infrared image fusion method in the framework of multi-scale decomposition. Adaptive iteration PCNN model is proposed which is used in visible and infrared image fusion combined the Contourlet and the PCNN theory. Then by simulation, we compare the fused performance of traditional PCNN and the proposed PCNN base on wavelet, down-sampling Contourlet and non-down-sampling Contourlet. The result shows, the new PCNN model can promote the fusion effect.In the second part of this article, the compressed sensing theory was applied to visible and infrared image. The reconstruction algorithm of visible and infrared image compressed sensing was researched emphatically and some usual algorithms were compared. At last the existing visible and infrared image fusion algorithms based on CS were compared and then the adaptive PCNN was applied to this framework for the first time. By simulation, we discuss the feasibility of using PCNN in the visible and infrared image fusion algorithm base on CS, and get a better fusion effect by this method.