Flavonoid Distribution Mapping System of Velvet Apple Leaf Based on Hyperspectral Imaging

Abstract

The total content of flavonoids in plants is generally measured using spectrophotometric analysis based on color absorption rates. The method could not inform the distribution of flavonoids in a leaf. In this study, the mapping system of flavonoid distribution of Velvet Apple (Diospyros discolor Willd.) leaf was introduced using a hyperspectral imaging technique combining spectral and spatial analysis. The proposed system consists of a measurement system and a mathematical model that converts each spatial pixel into a value that represents the number of flavonoids in velvet apple leaves. The measurement system consists of a hyperspectral camera, halogen lamp, slider, and measurement frame. A random forest (RF) method is used to calculate the transformation model between reflectance values and total flavonoids. The construction of the measurement system was carried out with 738 data containing spectral data and lab measurement data. The evaluation of the random forest model obtained a value of R2 of 0.94 and RMSE 15.87 mg/ml.