Advances in Very-High-Resolution Remote Sensing

T his special issue addresses very-high-resolution (VHR) remote sensing data processing and applications. As a matter of fact, satellite and airborne remote sensing is currently undergoing a technical revolution with the appearance and blooming development of VHR sensors. For these sensors, high resolution can be spectral with hundreds of narrow bands, spatial where the resolution of image pixels can be much less than 1 m, and temporal where the data are collected from the same area on multiple dates. The need for VHR remote sensing data has grown in parallel to this technical revolution. Application areas for VHR remote sensing images include energy (bio-mass energy and forest inventory, optimal solar photovoltaic installations , snow cover monitoring for the prediction of hydroelectric energy production), water (prevention and management of draughts, floods, monitoring of water quality, understanding of the oceanic circulation at mesoscales and smaller, observations of the temporal and spatial variations in water volumes stored in rivers, lakes, and wetlands in order to fulfill our basic need for fresh water), and the environment (detection of pollution, measurement of the climate change, monitoring of urban growth and management of urban planning, data assimilation with large-scale models). In order to fully exploit all the potential offered by the new generations of VHR sensors and to actually face all the applications with a very high societal impact, advanced signal and image processing methods are required. Unfortunately, most of the traditional processing algorithms fail when the resolution increases significantly. For instance, statistical learning becomes intractable with hyperspectral images because of the dimensionality of the data. Similarly, while it was easy to classify urban versus nonurban areas with medium resolution data, VHR data enable the accurate classification at the building scale, but this requires a complete redesign of the whole processing chain. These types of problems become more serious with synthetic aperture radar (SAR) because of its ''side-looking imaging geometry.'' There are also new multitemporal and multifre-quency fully polarimetric SAR systems This special issue focuses on the frontiers of very-high-resolution optical and microwave remote sensing data processing and on important application areas, including urban environments, precision agriculture, and natural disasters such as earthquakes.