This study addresses the challenge of efficiently evaluating sugarcane varieties for resistance to orange and brown rust using high-throughput phenotyping methods. Traditional visual evaluations by specialists are slow and subjective, hindering the development of desirable traits. Leveraging UAV-based multispectral data and machine learning algorithms, the research correlated spectral information with infection scores. The study successfully classified sugarcane varieties based on resistance using Random Forest, radial SVM, and KNN algorithms, achieving high accuracy levels. The findings highlight the potential of UAV-based multispectral data in expediting sugarcane phenotyping for enhanced disease resistance assessment.
