UAV IMAGERY AND MACHINE LEARNING: A DRIVING MERGER FOR BIOPHYSICAL MODELING OF SACCHARINE AND BIOENERGY FEEDSTOCKS IN SUGARCANE

• Grant number: 22/13992-0
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: April 01, 2023
• End date: March 31, 2024
• Field of knowledge: Agronomical Sciences - Agronomy - Crop Science
• Principal Investigator: Rouverson Pereira da Silva
• Grantee: Marcelo Rodrigues Barbosa Júnior
• Host Institution: Faculdade de Ciências Agrárias e Veterinárias (FCAV). Universidade Estadual Paulista (UNESP). Campus de Jaboticabal. Jaboticabal , SP, Brazil

Abstract

Estimating saccharine and bioenergy feedstocks in sugarcane allows stakeholders, from research centers to industries, to decide the precise time and place to harvest a product in the field; thus, it can streamline workflow while leveling the cost-effectiveness of full-scale production. On the one hand, Brix and Purity can offer meaningful and reliable indicators of high-quality raw materials for industrial food and fuel processing. On the other hand, cellulose, hemicellulose, and lignin are the main constituents of straw, directly contributing to bioethanol production. However, analyzing these materials in the laboratory is a time-consuming and non-scalable task. Therefore, we propose an approach based on multispectral UAV imagery, thermal and fluorescence data, with machine learning algorithms to develop a non-invasive and predictive framework to map crop quality indicators. Our study represents a breakthrough in the evaluation and monitoring of sugarcane on an industrial scale. Our insights can provide stakeholders with possibilities to develop resource-efficient, high- throughput, prescriptive harvesting lines for products and by-products.