Multi-source Image Fusion Algorithms Baesd On The NSCT

Multi-source image fusion is one of the important branch of information fusion, and also is an important technology of image processing and computer vision. Image fusion can generate a image that are more comprehensive, more accurate than a single sensor image. First, in this paper we introduce the basic theory of image fusion, then we focus on discussing the Nonsubsampled Contourlet Transform(NSCT), and then research fusion of the different types of image based on NSCT transform. The main research works are as follows:(1) According to the characteristics of the infrared and visible image, we propose an image fusion algorithm of infrared and visible images based on local average energy using NSCT. After the two registered original images are decomposed by using NSCT, the low frequency sub-band coefficients and band-pass directional sub-band coefficients of the fused image can be obtained by the fusion rules based on local average energy, and then the fused image is obtained by performing the inverse NSCT on the combined coefficients.(2) According to the characteristics of CT and MRI medical images, an adaptive fusion algorithm of CT and MRI medical images based on NSCT is presented in this paper. The source images are decomposed in a multi-direction way by using the nonsubsampled pyramids (NSP) and the nonsubsampled directional filter banks (NSDFBs). In band-pass directional sub-band coefficients fusion rules, we use local energy and the weighted average combination, meanwhile for the larger selection of absolute value of the coefficient are applied in the highest levels. The combination of the adjustable parameter and objective evaluation index of the adaptive fusion rules are used in low frequency sub-band fusion.The fusion simulations for several infrared images and visible images are taken, and the result analysis shows that the algorithm proposed in this paper is better than wavelet-based, Curvelet-based and NSCT-based method. The fusion experiments for several CT and MRI medical images are taken, and the result analysis shows that the algorithm proposed in this paper have a better integration of the information of the original image, a better fusion effect, and in accordance with the requirements of adaptive image fusion.