PREDICTION OF MOISTURE CONTENT IN CORN LEAVES BASED ON HYPERSPECTRAL IMAGING AND CHEMOMETRIC ANALYSIS

The moisture content of corn leaves under different moisture treatments was predicted with a hyperspectral imaging technique. The reflectance spectra of corn leaves were acquired in the spectral range of 900 to 1700 nm with a hyperspectral imaging system, and the moisture content of the same samples was acquired with the traditional oven-drying method. Partial least square regression (PLSR), multiple linear regression (MLR), and back-propagation artificial neural network (BP-ANN) models were used to analyze the correlation between spectral data and moisture content. Correlations were found between moisture content of the corn leaves and hyperspectral information; the coefficient of determination was greater than 0.85, with RMSEC and RMSEP values less than 0.01. This work provides an effective method for prediction of moisture content of corn leaves using a hyperspectral imaging technique.