Correcting the effects of field of view and varying illumination in spectral measurements of crops
2004
Steven, M.D.
Spectral measurements of crops offer valuable information on the state of the canopy for precision agriculture. Tractor-mounted sensors offer an effective means of providing this information because they can operate in all weathers, but any operational system must account for instrumental characteristics and for variations introduced by the weather. This study compared spectroradiometer measurements of field plots of winter-wheat made with a wide and a narrow instantaneous-field-of-view (IFOV), and with the crop canopy shaded to simulate the effect of cloud cover. Large and consistent differences in spectral reflectance were found with measurement configuration, particularly in the near-infrared: measurements made with a hemispheric IFOV were up to 60% higher on average than those with a narrow IFOV and measurements in shade were up to 40% higher than in bright sunlight. In spite of these large differences, IFOV and shading had no significant effect on the ability of vegetation indices to measure canopy density. A simple theoretical model was developed to separate the spectral dependence on leaf area from differences related to viewing and illumination angles. Fitted coefficients of the model were applied in an attempt to reconstruct the narrow IFOV sunlit spectrum from wide IFOV or shaded data. Spectra reconstructed from shade measurements were reasonably accurate, but reconstructions from wide IFOV were generally less accurate and failed drastically at low leaf area index. However, while the ability to correct for shade is essential to account for cloud cover variations during field surveys, the effect of IFOV is less critical and can be accounted for by recalibration of spectral relationships with canopy parameters.
Afficher plus [+] Moins [-]Mots clés AGROVOC
Informations bibliographiques
Cette notice bibliographique a été fournie par National Agricultural Library
Découvrez la collection de ce fournisseur de données dans AGRIS