Hyperspectral Image Analysis for Plant Stress Detection

Abstract

Plant stress significantly reduces plant productivity. Automated on-the-go mapping of plant stress allows for timely intervention and mitigation of the problem before critical thresholds are exceeded, thereby maximizing productivity. A hyperspectral camera analyzed the spectral signature of plant leaves to identify the plant water stress. Five different levels of water treatment were created on young apple trees (Buckeye Gala) in a greenhouse and continuously monitored with a hyperspectral camera along with an active-illuminated spectral vegetation sensor and a digital color camera. Individual spectral images over a 400 1000 nm wavelength range were extracted at a specific wavelength to estimate reflectance and generate spectral profiles for five groups of apple trees at different water treatment levels. Various spectral indices were investigated and correlated to stress levels. The highest correlation was found with Red Edge NDVI at 705 nm and 750 nm in narrowband indices and NDVI at 680 nm and 800 nm in broadband indices. The experimental results indicate that intelligent optical sensors could deliver decision support for plant stress detection and management.