Evaluation of a UAV-based hyperspectral frame camera for monitoring the leaf nitrogen concentration in rice

UAV based hyperspectral imaging is a promising approach to monitor crop growth status rapidly and non-destructively. This paper described a novel instrument to get hyperspectral information from lightweight unmanned aerial vehicles for crop monitoring. The objectives of this study were to assess the data quality of one hyperspectral frame camera and evaluate the ability in rice nitrogen status monitoring. In this study, we introduced one hyperspectral frame camera weighing 470g which could be mounted to low-weight UAVs (<;3 kg). The flight campaign was conducted in a paddy rice field in September 2015. During the flight, two ground-based portable spectrometers (ASD Field Spec Pro spectrometer and GreenSeeker RT 100) were used to collect rice canopy spectra. Later, Normalized difference vegetation index (NDVI) derived from hyperspectral images was compared with that from GreenSeeker and ASD. Also, field sampling was taken at the same day with the flight, and leaf nitrogen concentration (LNC) was obtained through Kjeldahl digestion method. Five existing vegetation indices that were used for N detection were used to estimate LNC. Results are satisfactory, which lay a foundation for the promising application of UAV-based hyperspectral remote sensing on precision agriculture.