Near-infrared Spectroscopy in detecting Leaf Miner Damage on Tomato Leaf

Automatic diagnosis of plant disease is important for plant management and environmental preservation in the future. The objectives of this study were to characterise the leaf reflectance spectra of tomato leaves damaged by leaf miner and to determine which wavelengths were most responsive to plant damage caused by the pest. Tomato leaf damage was classified into five scales based on the severity levels displayed on the surfaces of plant leaves. A spectral parameter of reflectance sensitivity was used to find the optimal wavelengths for determining and evaluating the damage level. Results from near-infrared spectroscopy showed that there were clear differences in spectral reflectance from different levels of infestation. Spectral reflectance decreases significantly with the increasing severity level at the short wavelengths of near infrared 800–1100 nm but changes for individual bands of 1450 and 1900 nm where spectral reflectance increases with the increasing severity level. Spectral parameters such as single-wavelength reflectance, peak area and water band index were used to discriminate the severity level of infestation. The results indicate that the sensitive bands of 1450 and 1900 nm modelled with severity level provided the highest correlation coefficient