eduzhai > Applied Sciences > Engineering >

Coupled Tensor Decomposition for Hyperspectral and Multispectral Image Fusion with Inter-Image Variability

  • Save

... pages left unread,continue reading

Document pages: 15 pages

Abstract: Coupled tensor approximation has recently emerged as a promising approach forthe fusion of hyperspectral and multispectral images, reconciling state of theart performance with strong theoretical guarantees. However, tensor-basedapproaches previously proposed assume that the different observed images areacquired under exactly the same conditions. A recent work proposed toaccommodate inter-image spectral variability in the image fusion problem usinga matrix factorization-based formulation, but did not account forspatially-localized variations. Moreover, it lacks theoretical guarantees andhas a high associated computational complexity. In this paper, we consider theimage fusion problem while accounting for both spatially and spectrallylocalized changes in an additive model. We first study how the generalidentifiability of the model is impacted by the presence of such changes. Then,assuming that the high-resolution image and the variation factors admit aTucker decomposition, two new algorithms are proposed -- one purely algebraic,and another based on an optimization procedure. Theoretical guarantees for theexact recovery of the high-resolution image are provided for both algorithms.Experimental results show that the proposed method outperforms state-of-the-artmethods in the presence of spectral and spatial variations between the images,at a smaller computational cost.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×