Image fusion is the process of combining relevant information from two or more images or sequence of the same scene delivered by different sensors simultaneously or asynchronously into a single highly informative image.In this paper, the current typical methods of image fusion are discussed. Then a new image fusion technology based on Orthogonal Matching Pursuit (OMP) and K singular value decomposition (K-SVD) is proposed based on the previous studies.The basic idea of the algorithm is to select a group of images which have similar structures with the source image as a sample sequence and train a redundant dictionary using K-SVD algorithm. Exploiting the resulted dictionary, the source images will be decomposed by the OMP algorithm. A set of atom vectors will be chosen from the trained dictionary, and then the source image will be expressed as a linear combination of these vectors.Compared with traditional fusion algorithms, the proposed algorithm can avoid the block-effect and ripple noise effectively and make the fusion image robust. It improves the quality of fusion image in a certain degree. The proposed algorithm can be applied in image analysis and computer vision field. |