Precision monitoring of rice nitrogen fertilizer levels based on machine learning and UAV multispectral imagery

Ming-Der Yang, Yu-Chun Hsu, Yi-Hsuan Chen, Chin-Ying Yang, Kai-Yun Li

Abstract: Highlights•Developing a UAV-based machine learning (ML) workflow for rice nitrogen (N) level classification.•Combining composite features and ML models to achieve 90 % overall accuracy.•Optimizing SVM performance with just four selected features via the Chi-square test.•Highlighting canopy coverage and NIR-related features for N level detection.•Assisting in reducing fertilizer wastage and emissions, reinforcing eco-friendly cultivation.