Characterization of vineyard training systems based on remote sensing and crop indices

In the context of precision viticulture, this work presents the implementation of remote sensing techniques to analyze the spatial variability of a vineyard (Vitis vinifera L.). This work seeks to continue a preliminary investigation conducted in 2020; this time, the study area within the vineyard w...

Descripción completa

Detalles Bibliográficos
Autores principales: Capraro, Flavio, Pacheco, Daniela, Campillo, Pedro
Formato: info:ar-repo/semantics/documento de conferencia
Lenguaje:Español
Publicado: IEEE 2025
Materias:
Acceso en línea:http://hdl.handle.net/20.500.12123/23992
https://ieeexplore.ieee.org/document/10735888
https://doi.org/10.1109/ARGENCON62399.2024.10735888
Descripción
Sumario:In the context of precision viticulture, this work presents the implementation of remote sensing techniques to analyze the spatial variability of a vineyard (Vitis vinifera L.). This work seeks to continue a preliminary investigation conducted in 2020; this time, the study area within the vineyard was expanded, and the campaigns of 2023 and 2024 were considered. This trial was conducted in a vineyard located in the province of San Juan, Argentina. The vineyard was divided into three blocks (replicates), and within each block, three training systems were randomly implemented: Free Cordon, Minimal Pruning and Box Pruning. The analysis was mainly based on extracting information from various vineyard maps constructed from high-resolution (2.5 cm pixel size) multispectral and thermographic images. These images were captured using special cameras mounted on an unmanned aerial vehicle (UAV). Vegetation indices NDVI and NDRE were calculated from the orthomosaics. The spatial distribution of each index and the crop temperature (Tc) were studied, and measurements were subsequently recorded in plants within each training system. Based on these measurements, significant differences were identified among the three training systems. The results demonstrated the usefulness of the high-resolution images acquired to assess the vineyard's condition at the plant level, allowing the producer to manage each training system specifically.