Identifying Individual Nutrient Deficiencies of Grapevine Leaves Using Hyperspectral Imaging

The efficiency of a vineyard management system is directly related to the effective nutritional disorders, which significantly downgrades vine growth, crop yield and wine quality. To detect we successfully extracted wide range features using hyperspectral (HS) images identify healthy individual nutrient deficiencies grapevine leaves. Features such as mean reflectance, first derivative variation index, spectral ratio, normalised difference vegetation index (NDVI) standard deviation (SD) were employed at various stages in ultraviolet (UV), visible (VIS) near-infrared (N.I.R.) regions for our experiment. Leaves examined visually laboratory grouped either (i.e. control) or unhealthy. Then, leaves from these two groups. In second experiment, nutrient-deficient (e.g., N, K Mg) also analysed compared with those control Furthermore, customised support vector machine (SVM) was used demonstrate that can be utilised high degree effectiveness unhealthy samples not only distinguish deficient but defects. Therefore, proposed work corroborated HS imaging has excellent potential analyse based on healthiness