papers AI Learner
The Github is limit! Click to go to the new site.

Blind Hyperspectral-Multispectral Image Fusion via Graph Laplacian Regularization

2019-02-21
Chandrajit Bajaj, Tianming Wang

Abstract

Fusing a low-resolution hyperspectral image (HSI) and a high-resolution multispectral image (MSI) of the same scene leads to a super-resolution image (SRI), which is information rich spatially and spectrally. In this paper, we super-resolve the HSI using the graph Laplacian defined on the MSI. Unlike many existing works, we don’t assume prior knowledge about the spatial degradation from SRI to HSI, nor a perfectly aligned HSI and MSI pair. Our algorithm progressively alternates between finding the blur kernel and fusing HSI with MSI, generating accurate estimations of the blur kernel and the SRI at convergence. Experiments on various datasets demonstrate the advantages of the proposed algorithm in the quality of fusion and its capability in dealing with unknown spatial degradation.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1902.08224

PDF

http://arxiv.org/pdf/1902.08224


Similar Posts

Comments