eduzhai > Applied Sciences > Engineering >

Spectral Superresolution of Multispectral Imagery with Joint Sparse and Low-Rank Learning

  • king
  • (0) Download
  • 20210505
  • Save

... pages left unread,continue reading

Document pages: 12 pages

Abstract: Extensive attention has been widely paid to enhance the spatial resolution ofhyperspectral (HS) images with the aid of multispectral (MS) images in remotesensing. However, the ability in the fusion of HS and MS images remains to beimproved, particularly in large-scale scenes, due to the limited acquisition ofHS images. Alternatively, we super-resolve MS images in the spectral domain bythe means of partially overlapped HS images, yielding a novel and promisingtopic: spectral superresolution (SSR) of MS imagery. This is challenging andless investigated task due to its high ill-posedness in inverse imaging. Tothis end, we develop a simple but effective method, called joint sparse andlow-rank learning (J-SLoL), to spectrally enhance MS images by jointly learninglow-rank HS-MS dictionary pairs from overlapped regions. J-SLoL infers andrecovers the unknown hyperspectral signals over a larger coverage by sparsecoding on the learned dictionary pair. Furthermore, we validate the SSRperformance on three HS-MS datasets (two for classification and one forunmixing) in terms of reconstruction, classification, and unmixing by comparingwith several existing state-of-the-art baselines, showing the effectiveness andsuperiority of the proposed J-SLoL algorithm. Furthermore, the codes anddatasets will be available at:this https URL TGRS J-SLoL, contributing to the RScommunity.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×