Image Fusion Technique Using Singular Value Decomposition Based Undecimated Subband Decomposition

Image fusion is an important area of research. In remote sensing, the usage of the same sensor in different working modes, or different image sensors, can provide complementary information. Therefore, it is needed to fuse outputs from multiple sensors (or the same sensor in different working modes) to improve the performance of the remote images, which are very useful for many image processing tasks. Singular Value Decomposition Undecimated Subband Decomposition (SVD-USBD) of images with its Multi-Resolution schemes and based on it novel image fusion technique have been presented. The performance of this algorithm is compared with that of often used image fusion technique using Multi-Resolution form of Singular Value Decomposition (MR-SVD). It is observed that image fusion by SVD-USBD provides almost similar fusion quality and that is why it can be used in combinations with other methods to develop hybrid schemes for multi-focus image fusion or fusion of different types of images as infrared or termal imaging.