Work

Comparison of NDVI obtained from an active proximal sensor and UAV multispectral imagery in coffee

Year: 2026

Type: preprint

Abstract: The Normalized Difference Vegetation Index (NDVI) has been widely used in coffee crop monitoring. The direct comparison between NDVI measurements obtained using different methods is important for data integration and the construction of time series with information from various crop monitoring platforms. In this study, the NDVI obtained from the GreenSeeker active sensor and obtained via images from a multispectral camera coupled to an unmanned aerial vehicle (UAV) was compared at different phenological stages of the coffee crop. The study was conducted in three Coffea arabica L. crops in the Matas de Minas region. NDVI data were collected by UAV and by GreenSeeker in 14 fields distributed across Paula Cândido, Fervedouro, and São Sebastião da Vargem Alegre, during the 2024/2025 and 2025/2026 harvest seasons. The evaluations were carried out at the pinhead, expanding, and maturation stages, between November and April of each harvest season. The NDVI values obtained by UAV and by GreenSeeker were compared at each phenological stage using bias, RMSE, and paired t-test (p < 0.05). The agreement between the methods was also evaluated using Bland-Altman analysis, with calculation of bias and 95% limits of agreement, which allowed verification of the absence of possible systematic deviations between the measurements. The results suggest agreement between the NDVI acquisition methods at all phenological stages, with no significant statistical difference observed. The low bias and RMSE values, together with the Bland-Altman analysis, confirm the consistency and equivalence between the methods in estimating NDVI in coffee crops, regardless of the phenological stage evaluated.

Source: Research Square

DOI: 10.21203/rs.3.rs-10767247/v1