Early Identification of Plant Stress in Hyperspectral Images

In recent years, remarkable results have been achieved in the early detection of weeds, plant diseases and insect pests in crops. These achievements are related both to the development of non-invasive, high resolution optical sensors and data analysis methods that are able to cope with the resolution, size and complexity of the signals from these sensors. Especially hyperspectral cameras are capable sensors for the early detection of stress even before visible sympotoms become apparent. Their interpretation with regard to data amount, noise factors and their unknown effects is challenging. In this study, the the focus lies on the early detection of drought stress in barley plants based on hyperspectral images. For the specific task of representing and predicting the development of drought stress different model types are compared. It turns out that the linear ordinal classification combines both, high precision and low model complexity. Prediction results for an drought stress experiment over time are presented.