A novel image fusion method based on orthonormalization transform combined with wavelet

With the availability of multi-sensor, multi-temporal, multi-resolution and multi-frequency image data from operational Earth observation satellites the fusion of digital image data hits become a valuable tool in remote sensing, imagge evaluation. Although most image fusion methods based on transform try to construct an orthonormalized space, the physical meaning of the transform is not always clear enough. As high resolution images and low resolution images are all the record of the same real world, more attention should be paid on the physical meaning of these data and the transform on them. Still, present pixel level image fusion methods ignore spatial information or spectral correlation, which is a bit, problem to achieve reasonable results. The paper analyzes the mechanics of the image fusion methods based on orthonormalization transform, and brings up a novel image fusion method based on orthonormalization transform combined with wavelet transform, which is much more physically meaningful and keep more spatial and spectral information.