What question did this study set out to answer?

This study aims to evaluate the effectiveness of multispectral and hyperspectral data in estimating coffee yield using machine learning algorithms.

July 10, 2026Open Access

Multiscale Multispectral–Hyperspectral Data for Estimating Coffee Yield Using Machine Learning Algorithms

GMGeorge Deroco MartinsUniversidade Federal de Uberlândia LCLucas Henrique Vicentini Viana de CarvalhoUniversidade Federal de Uberlândia FSFilipe Vieira da SilvaUniversidade Federal de Uberlândia

Key Points

This study aims to evaluate the effectiveness of multispectral and hyperspectral data in estimating coffee yield using machine learning algorithms.
Evaluated multispectral (Mavic 3M) and hyperspectral (Blue Wave) data across 40 different coffee plots.
Used linear regression, support vector machines (SVM), and neural networks for analysis.
Compared correlation coefficients to assess data performance in estimating crop productivity.
Neural network models using hyperspectral data achieved the highest performance (r = 0.92; RMSE = 6.6%)
Multispectral models showed lower performance (r = 0.84; RMSE = 9.4%).
Overall, multispectral data exhibited strong correlation with productivity, especially in the Red Edge and Green bands.

Abstract

Abstract. This study evaluated the performance of multispectral (Mavic 3M) and hyperspectral (Blue Wave) data in estimating coffee crop productivity using linear regression, SVM, and neural networks. Forty plots with different varieties were analyzed. Multispectral data showed high correlation with productivity, especially the Red Edge (r = 0.704) and Green (r = 0.644) bands. For hyperspectral data, PRI (r = 0.535), GNDVI (r = -0.394), NDVI (r = -0.33), and CIRE (r = -0.328) were significant, highlighting the negative correlation pattern typically observed in perennial crops. Neural network models applied to hyperspectral data achieved the best performance (r = 0.92; RMSE = 6.6%), surpassing multispectral models (r = 0.84; RMSE = 9.4%).

AI에게 질문

Bookmark

View Full Paper