Abstract
The fusion of multisensor data has attracted a lot of attention in computer vision, particularly among the remote sensing community. Hyperspectral image fusion consists in merging the spectral information of a hyperspectral image with the geometry of a multispectral one in order to infer an image with high spatial and spectral resolutions. In this paper, we propose a variational fusion model with a nonlocal regularization term that encodes patch-based filtering conditioned to the geometry of the multispectral data. We further incorporate a radiometric constraint that injects the high frequencies of the scene into the fused product with a band per band modulation according to the energy levels of the multispectral and hyperspectral images. The proposed approach proved robust to noise and aliasing. The experimental results demonstrate the performance of our method with respect to the state-of-the-art techniques on data acquired by commercial hyperspectral cameras and Earth observation satellites.
Highlights
Image fusion has been an active field of research due to the growing availability of data and the need of gathering information from different imaging sources [1,2]
Since the ground truth images are available, we evaluate the results in terms of the root mean squared error (RMSE), which accounts for spatial distortions, and the spectral angle mapper (SAM), which measures the spectral quality
We have presented a convex variational model for HS and MS data fusion based on the image formation models
Summary
Image fusion has been an active field of research due to the growing availability of data and the need of gathering information from different imaging sources [1,2]. The bandwidth capacity as well as the onboard storage are important limiting factors too These tradeoffs lead to acquiring a multispectral (MS) image with high spatial but low spectral resolution, or an HS image with accurate spectral but poorer spatial resolution. The difference between the two is that the geometry is encoded in a grayscale image in the case of pansharpening, and that the number of spectral bands to be spatially interpolated is much lower than in HS fusion. The proposed model is compared with several state-of-the art techniques on both remote sensing imagery and data captured by HS cameras of everyday-life scenes.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
