Early detection of the fungal disease "apple scab" using SWIR hyperspectral imaging

The aim of our study is to examine the potential of SWIR hyperspectral imaging to early detect apple scab infection. Close range hyperspectral images of healthy and infected leaves were acquired daily under laboratory conditions from 2 days to 11 days after inoculation using a push-broom SWIR camera. A PLS-DA classification model was built at the advanced infection stage D11 and was applied on the infected and healthy leaves images acquired at others infection stages. This study showed that good predictions can be achieved when classifying infected leaf regions based on hyperspectral data using PLS-DA. Results suggest that the spectral domain between 1000 - 2500 nm is suited to early differentiation between infected and healthy leaves. At early infection stages, the water absorption band at 1940 nm has the major discriminatory effect.