Sugarcane Stalk Content Prediction in the Presence of a Solid Impurity Using an Artificial Intelligence Method Focused on Sugar Manufacturing

Abstract

For the first time in literature, an analytical method was developed using artificial neural networks (ANNs) combined with color information from digital images to predict the content of sugarcane stalks in the presence of a solid impurity. The data were generated using a laboratory-made simple imaging system and free-access computational routine for the conversion of the images into 10 colors. The ANN model was implemented using 10 neurons in the input layer, 8 neurons in the hidden layer and 1 neuron in the output layer related to the content of sugarcane stalks. The ANN model provided relative errors of 3% and achieved correlation coefficients of 0.98, 0.93, and 0.91 for the training, validation and test sets, respectively. A partial least squares (PLS) model showed the nonlinear nature of the data that implies the application of ANN model. The developed method has the potential to be applied in sugarcane mills as an improvement for the production of high-quality sugar.