Differentiation of Hard Red Wheat by Near-Infrared Analysis of Bulk Samples

Abstract

most often, about 95% for samples grown during the same years as used in calibration. These accuracies were significantly better than those associated with discriminant models that were based solely on protein content, NIR-hardness, or a combination of protein and hardness. Spectrally sensed water-matrix interactions were probably beneficial to model accuracy; however, moisture content alone was not deemed necessary to a model's success. When predicting the fourth year, the MLR model needed a bias correction, whereas the other three models performed reasonably well. The ANN model's performance was highest, with accuracies in the range of 95-98%. At little expense to model accuracy, the number of input nodes to the ANN model could be reduced from 223 to Ill, provided the full wavelength range was preserved. to mean protein and hardness levels would be necessary. The current study differs from the previous one in that examination by NIR is performed on bulk wheat without first grinding the sample; however, the same samples that constituted the calibration, validation, and prediction sets have been used. If successful, classification by NIR spectroscopy would be advantageous over digital imaging in terms of equipment cost and computational processing time. The objectives of the current study were to develop accurate models for the differentiation of HRW and HRS wheats based on NIR reflectance spectra of bulk samples and to compare various classification algorithms. Although the scope of the study was limited because individual kernels were not examined and, hence, detection of mixtures of classes was not possible, the study represented the first comprehensive attempt to determine whether differences in intrinsic properties of the hard red wheat classes are measurable by NIR spectroscopy.