To enhance the spatial resolution of VRE bands of Sentinel-2 image and perform a jointed analysis with Pléiades-1 images, the renowned small-data learning theory, convex/deep (CODE), is employed, forming a completely unsupervised high-quality synthesis system with fast closed-form implementation. More specifically, we formulate this synthesis problem as a satellite image fusion problem (i.e., Sentinel-2 and Pléiades-1 images). Initially, both images are fused by the proposed unsupervised Deep Residual Convolution Network (DRCN). The remarkable AI property, Deep Image Prior (DIP), enables DRCN to generate a preliminary rough solution entailing feasible spectral characteristics. Subsequently the convex Q-quadratic norm is leveraged to bridge the convex optimization and deep learning. Eventually, a powerful convex solver, the Alternating Direction Method of Multipliers (ADMM), is employed for solving the convex problem; this is the essence of CODE theory. Owing to the superiority of small-data learning technology, the high-quality synthesis of 4m/2.5m image products in multiple VRE bands for Sentinel-2 data can be acquired in a completely unsupervised manner.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.