Determination of the corn hotspot diurnal and seasonal change by real scene models

Abstract

The corn hotspot dependence on its structure at various times in one day and various growth periods is investigated. We simulate the scattering of light in the canopy based on three dimensional corn canopies. Our approach consists of seven processes. Firstly, the biophysical parameters of corn, such as plant height, leaf base height, size of leaf, and shape of leaf, are measured on field, and the statistics of the parameters is obtained based on the measurement. Secondly, the three dimensional structure of single corn is re-built by the L-system and the statistics. Thirdly, the real pixel scene is configured according to the spatial pattern of corn in pixel, such as planting by row or by random distribution. Fourthly, hotspots are calculated by a light interception model under various viewing geometric conditions. Fifthly, the correlation of corn structure parameters and diurnal consecutive hotspots is evaluated, and key factors are sought. Sixthly, the correlation of corn structure parameters and seasonal consecutive hotspots also is evaluated, and key factors are sought. Lastly, the dependence of the key factors that form seasonal consecutive hotspots on the ones that form diurnal consecutive hotspots is estimated at various growth periods. We obtain key factors concerning consecutive hotspots. The result will improve our understanding about hotspots in Triana satellite data, and help us to use the data effectively.