Assessment of crop health in relation to competition from weeds: using balloon-borne images and pectrometer data

Remote sensing is increasingly being applied to agricultural applications. It offers the ability to quickly and efficiently supply information about the spatial variability that occurs within fields. Recently remote sensing technology has been successfully applied to identify and mapping weeds within crops. Weeds pose a serious threat to crop health and can adversely affect yield. The opportunity exists to explore new applications of remote sensing in the agricultural setting particularly measuring crop health. This study covers the application of remote sensing to detect weed-induced stress in grain crops. The aim of the study was to determine the ability of spectral data, captured by a low cost balloon-borne camera system and a spectroradiometer, to discriminate between a healthy and stressed crop. Spectral data was collected from a weed trial site that contained several sorghum varieties planted at different planting densities and with or without weeds. Discriminant function analysis was conducted on the data captured by the two sensors to analyse the ability of spectral data to be used to differentiate between the crop containing weeds and the crop without weeds. The results of the analysis indicate that the spectral data from both sensors can be used to distinguish between the healthy and stressed crop. These results provide sufficient cause to suggest further research is warranted into the application of remote sensing as a tool for measuring crop health