Study On Key Technology Of Infrared And Visible Image Fusion

As an important branch of multi-source image fusion, infrared and visible image fusion involves the basic theory and methods of image fusion, and also has its own features. Infrared imaging is based on the thermal radiation of scene, and it is not susceptible to weather and illumination. But infrared images are rather vague as a whole, and the spatial resolution and image contrast of infrared images are low. In contrast, visible imaging is based on the reflection of visible light. The spatial resolution of visible images is higher, and they have clear texture information and abundant details. However, they are vulnerable to the interference of illumination and climatic conditions. Therefore, it has great significance to do further research on how to fuse infrared and visible images with their respective characteristics. So this paper conducts an intensive study on the key technology of infrared and visible image fusion.Firstly, the basic theory, principle, framework and the main methods are studied, which lays the theoretical foundation for the study of infrared and visible image fusion technology.Then the imaging principle and image characteristics of infrared and visible image are analyzed and compared. And the subjective quality evaluation and objective quality evaluation are described in this paper.This paper carries thorough research on principle and implementation method of non-sampled Contourlet transform(NSCT), for multi-scale analysis tools is one of the most widely theory applied in the image fusion field. And this multi-scale analysis tool is used as the basic tool of the image fusion method in this paper. Based on this, infrared and visible image fusion algorithms based on visual saliency and robust principal component analysis(RPCA) are proposed respectively. Visual saliency is a reflection of the ability of target in the scene to attract visual attention of people. This characteristic is consistent with that infrared image and visible image fusion need to extract the target information of the infrared image, so this paper proposes a new infrared and visible image fusion algorithm based on visual saliency and NSCT. In addition, RPCA supposes that a data matrix can be represented as the superposition of a low-rank component and a sparse component. To apply the theory of RPCA to image processing field, this paper puts forward another infrared and visible image fusion based on RPCA and NSCT.The proposed visual saliency based fusion algorithm and RPCA based fusion algorithm are compared to benchmark algorithms by experiment. By assessed by subjective and objective quality evaluation, the experimental results verify that both of the proposed methods can not only highlight targets information in infrared images, but also take advantage of the abundant background information in visible images.