SINGLEWHEAT KERNEL COLOR CLASSIFICATION USING NEURAL NETWORKS

An optical radiation measurement system, which measures reflectance spectra from 400 to 2000 nm, was used to quantify single wheat kernel color. Six classes of wheat were used for this study. A neural network (NN) using input data dimension reduction by divergence feature selection and by principal component analysis was used to determine single wheat kernel color class. The highest classification accuracy was 98.8% when the divergence feature selection method was used to reduce the number of NN inputs. The highest classification accuracy was 98% when principal component analysis method was used to reduce the number of NN inputs.