Development and Calibration of Sensors for UAVs.

The unmanned airborne vehicle (UAV) based remote sensing with lightweight multi- and hyperspectral imaging sensors offer low-cost tools for the agricultural applications. Based on the accurate measurements of the way in which the vegetation reflect and emit energy, wide range of variables that affect the crops can be monitored. One of the very interesting recent innovations in the field of lightweight UAV imaging sensors is the area format hyperspectral imaging based on Fabry-Perot interferometer (FPI). With the FPI spectral camera, it is possible to collect area format spectral data cubes with stereoscopic and multiview setups, and to provide 3D surface information based on stereoscopy; this is expected to be very powerful measurement technology in the agricultural applications. Condition for reliable applications is high quality input data. Important step in the data processing is to provide data characterizing the physical reflectance characteristics of objects. In typical projects, the area of interest is covered by hundreds or thousands of spatially overlapping images collected under variable illumination conditions. Prerequisite for the data processing is the accurate sensor calibration in laboratory. Information about the illumination conditions is collected during the flight using ground and UAV based irradiance instruments. After data collection, the laboratory calibration is applied to the images. Geometric processing includes the image orientation and digital surface model generation. Then radiometric post-processing takes place to provide reflectance image mosaics and point clouds; the processing includes elimination impacts of varying atmospheric conditions and other instabilities, elimination of radiometric differences due to effects of view/illumination geometry, and the reflectance transformation. In this presentation, we will describe the new spectrometric UAV imaging system based on FPI spectral camera and the rigorous data processing chain. Finally, we will present results of several case studies, including precision agriculture, tree canopy insect damage assessment and water quality monitoring.