Remote sensing and GIS group research activities are based on derivation of useful information aggregated from various remote sensors. Remote sensing is based on satellite imaging as well as on images from color, multispectral, hyperspectral and thermal cameras mounted on robots or low-altitude UAVs – all available in the Remote sensing and GIS laboratory of the Institute.
The group’s research addresses processing, storage and retrieval of data acquired from multimodal sensors, and integration of large amounts of multimodal data acquired from different sources (such as WSNs, satellites, meteorological measurement stations, field surveys, etc.). The group’s activities also include the development of systems for instant access to relevant data presented in ways which are the most informative to end-users, such as GIS databases. Imaging techniques are combined with state-of-the-art machine learning approaches such as support vector machines, deep learning, boosting, evolutionary algorithms, and ensemble classification, and then applied in many innovative solutions for decision making, such as systems for crop detection, yield estimation or smart irrigation.
When it comes to data acquisition, the group possesses different imaging instruments, including several types of multispectral cameras, polarization filters, three hyperspectral cameras, high resolution thermal camera, lenses, high frame rate industry cameras, WSN kits, but also different acquisition platforms such as small autonomous UAV’s, motorized paragliders and autonomous unmanned ground vehicles.
Thermal and multispectral imaging of plants and crops using in-house available UAVs