MULTIVARIATE ANALYZING AND ARTIFICIAL NEURAL NETWORKS FOR PREDICTION OF PROTEIN CONTENT IN WINTER WHEAT USING SPECTRAL CHARACTERISTICS
This study aimed to predict the protein content(PC) and canopy spectra in winter wheat were measured based on field test. Key spectral bands were chosen by principal component analysis (PCA) method, and the predicted models were built by Partial Least Squares Regression (PLSR) and Artificial Neural Network (ANN). The performance of the feed forward and cascade forward ANNs was compared with those of PLS regression models using root mean square error (R) and the correlation coefficient (2). The finest consequence by CFBP was related to topology of 8-8-1 with Levenberg-Marquardt (LM) algorithm, threshold function of TANSIG-TANSIG-PURELIN and the initial strategy. This arrangement output was R=0.0289 and 2=0.9881 at 14 epochs. The consequences of estimate for correlation values of PLSR model was 0.9783. The results of prediction for the two models were in order of ANN > PLSR with correlation values of 0.9881 and 0.9783, respectively. Therefore, NIRS shows the potential for predicting protein content with accuracies suitable for process control.