Abstract
Convolutional Neural Networks (CNNs) consistently proved state-of-the-art results in image Super-resolution (SR), representing an exceptional opportunity for the remote sensing field to extract further information and knowledge from captured data. However, most of the works published in the literature focused on the Single-image Super-resolution problem so far. At present, satellite-based remote sensing platforms offer huge data availability with high temporal resolution and low spatial resolution. In this context, the presented research proposes a novel residual attention model (RAMS) that efficiently tackles the Multi-image Super-resolution task, simultaneously exploiting spatial and temporal correlations to combine multiple images. We introduce the mechanism of visual feature attention with 3D convolutions in order to obtain an aware data fusion and information extraction of the multiple low-resolution images, transcending limitations of the local region of convolutional operations. Moreover, having multiple inputs with the same scene, our representation learning network makes extensive use of nestled residual connections to let flow redundant low-frequency signals and focus the computation on more important high-frequency components. Extensive experimentation and evaluations against other available solutions, either for Single or Multi-image Super-resolution, demonstrated that the proposed deep learning-based solution can be considered state-of-the-art for Multi-image Super-resolution for remote sensing applications.
Highlights
Super-resolution (SR) algorithms serve the purpose of reconstructing high-resolution (HR) images from either single or multiple low-resolution (LR) images
Multi-image SR (MISR) aims at recovering an HR image IHR from a set of T LR images I[L1R,T] of the same scene acquired in a certain temporal window
The computation of the corrected mean squared error (cMSE) is performed in the same way as we did for the loss in Equation (12): it is the minimum MSE between IcSrRop + bu,v and the HR patches IuH,vR: cMSE = min u,v∈[0,6]
Summary
Super-resolution (SR) algorithms serve the purpose of reconstructing high-resolution (HR) images from either single or multiple low-resolution (LR) images. Due to constraints such as sensor limitations and exceedingly high acquisition costs, it is often challenging to obtain HR images. In this regard, SR algorithms provide viable opportunity to enhance and reconstruct HR images from LR images recorded by the sensors. Onboard instruments on satellites continue to generate ever-increasing data as spatial and spectral resolutions increase, and this has progressively become challenging for compression algorithms [1], as they try to meet the bandwidth restrictions [2,3]. It is imperative that enhancements and progress be made in post-processing techniques to overcome obstacles of increasing spatial resolution
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: 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.