2026-10-13T17:59:27Z https://www.scidb.cn/oai

10.57760/sciencedb.17395 2024-04-01T11:15:04Z

2024-04-01 SEN2NAIP: A large-scale dataset for Sentinel-2 Image Super-Resolution doi:10.57760/sciencedb.17395 en The increasing demand for high spatial resolution in remote sensing imagery has led to the necessity of super-resolution (SR) algorithms that convert low-resolution (LR) images into high-resolution (HR) ones. To address this need, we introduce SEN2NAIP, a large remote sensing dataset designed to support conventional and reference-based SR model training. SEN2NAIP is structured into two components to provide a broad spectrum of research and application needs. The first component comprises a cross-sensor dataset of 2,851 pairs of LR images from Sentinel-2 L2A and HR images from the National Agriculture Imagery Program (NAIP). Leveraging this dataset, we developed a degradation model capable of converting NAIP images to match the characteristics of Sentinel-2 imagery (S2like). Subsequently, this degradation model was utilized to create the second component, a synthetic dataset comprising 17,657 NAIP and S2like image pairs. With the SEN2NAIP dataset, we aim to provide a valuable resource that facilitates the exploration of new techniques for enhancing the spatial resolution of Sentinel-2 satellite imagery. remote sensing ; Sentinel-2; deep learning; super resolution Cesar Aybar David Montero Julio Contreras Simon Donike Freddie Kalaitzis Luis Gómez-Chova PUBLIC https://creativecommons.org/licenses/by/4.0/ dataset Science Data Bank