Using hyperspectral remote sensing as a tool for early detection of leaf rust in blueberries

Abstract

Monitoring and early detection of leaf rust caused by fungus 'Naohidemyces vaccinii' in plants is critical for sustainable production of blueberries in Australia. In this study, the main aim was to use hyperspectral remote sensing as a tool to detect leaf rust at an early stage in blueberries. Reflectance was measured in the wavelength range from 350 to 2500nm using a handheld hyperspectral spectro-radiometer. Differences in spectral reflectance were seen at a number of wavelengths in the visible, NIR and SWIR regions for the Sharpblue variety under field conditions. In glasshouse conditions the OB1 and Sharpblue varieties showed differences between inoculated and uninoculated plants in NIR and SWIR regions. The NIR region showed significant spectral difference between the three varieties of blueberry. The results indicate the possibility to detect differences in healthy and leaf rust infected blueberry plants at an early stage of the diseases using hyperspectral remote sensing techniques.