Use of artificial neural networks for the estimation of impurities in mixtures of sugarcane, soil and plant material

• Grant number: 18/03690-1
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2018
• End date: June 30, 2020
• Field of knowledge: Interdisciplinary Subjects
• Principal Investigator: Erica Regina Filletti Nascimento
• Grantee: Lucas Janoni dos Santos
• Host Institution: Instituto de Química (IQ). Universidade Estadual Paulista (UNESP). Campus de Araraquara. Araraquara , SP, Brazil

Abstract

We propose the development of Artificial Neural Networks (RNAS) as an alternative tool to estimate the fraction of impurities in mixtures of sugarcane, soil and plant material, from digital images of samples of these mixtures in different proportions. The ability to learn through examples and generalize the information learned is undoubtedly the main attraction of problem solving through ANNs. Generalization, which is associated with the ability of the neural network to learn through a reduced set of examples and then give coherent answers to unknown data, is a demonstration that the ability of ANNs goes far beyond simply mapping input ratios and output. Due to these advantages, this tool will be very useful for the study of impurities in the sugarcane, and may bring a possible solution for the characterization of the raw material delivered to the mills. (AU)