Imagens aéreas para análise da cobertura vegetal da pastagem

Abstract

The objective was to evaluate pasture vegetation cover over two agricultural years using vegetation indices from RGB aerial images obtained with drones and multispectral satellite imagery. The study area was already divided into six pasture paddocks, with three paddocks of Megathyrsus maximum cv. Mombaça and three of Urochola brizantha cv. MG-4. Data collection took place in January and July of 2022/2023. The study followed the principles of precision agriculture, with sampling grids tailored to each pasture, averaging approximately 4 samples per hectare. Field data on height, percentage of exposed soil, chlorophyll, and green biomass were collected. RGB aerial images captured by drones and multispectral images from Sentinel 2A satellite were obtained at four different times. By calculating vegetation indices such as Normalized Difference Vegetation Index (NDVI), Green/Red Vegetation Index (GRVI), Redness Index (RI), and Green Leaf Index (GLI), pasture vegetation cover (CVP) was determined. Results of CVP indicated slightly degraded pastures in the years 2022/2023. UAV (Unmanned Aerial Vehicle) images aided in identifying specific location failures with greater detail, which guided optimal pasture management practices.