HYPERSPECTRAL/MULTISPECTRAL IMAGER FOR MICROSATELLITES WITHIN-SITU DATA COLLECTION FOR PLANT PRODUCTION MONITORING ANDMODELLING

摘要

A joint South African – Flemish research project isdeveloping an integrated hyperspectral/ multispectralimager with in-situ data capturing capability to gathercritical inputs for self-learning models of capitalintensiveplant production systems.Optical remote sensing on this multi-sensor microsatelliteimager is provided by 3 classes of detectors, allon the same focal plane of a single telescope:a) 2 detectors providing 6 Multi-spectral bands inVNIR at 4.6 m GSD and a panchromatic detectorwith 2.7 m GSD.B) Hyperspectral detectors: 200 narrow spectral bands,over the range of 400 nm to 2350 nm.C) Video detector that provides snapshots andcontinuous video images at 3.3 m GSD.Ongoing research with these multi- and hyperspectraldata types deals with the early detection of abiotic andbiotic stress in orchards, bi-directional reflectancedistribution functions of vegetative material, modellingof crop production systems and incorporation of in-situmeasurements from sensors located in orchards.