Ricardo Vilar Neves

Purpose: Unmanned aerial vehicle (UAV)-based remote sensing can support non-destructive and spatial pasture monitoring, particularly in heterogeneous tropical systems. This study assessed the potential of UAV-derived vegetation indices (VIs) from the visible and near-infrared (NIR) spectrum to estimate canopy height and leaf area index (LAI) of ‘Tifton 85’ bermudagrass under conditions characterized with high spatial variability. Methods: Mean canopy height and LAI were measured at 55 georeferenced sampling points during five field campaigns between May 15 and June 22, 2023. Multispectral UAV imagery was used to derive 10 VIs. Relationships between VIs and observed variables were assessed using Spearman’s correlation coefficient (R _s ). Simple nonlinear models were fitted and assessed using the coefficient of determination (R²) and the standard error of estimate (SEE), and tested with an independent dataset using R², root mean square error (RMSE), and the modified Willmott index (d _mw ). Results: Near-infrared (NIR)-based indices showed the strongest relationships with pasture parameters. Similar to LAI, height was strongly correlated with some VIs (R _s > 0.80) and good fit to the nonlinear models (R² > 0.64). However, height model performance testing was poor (R² < 0.25, d _mw < 0.49, and nRMSE > 40%) and lower than that obtained for LAI (R² < 0.72, d _mw < 0.64, and nRMSE > 22%). Overall, the best performances were obtained with polynomial models using Normalized Difference Vegetation Index (NDVI) or Visible Atmospherically Resistant Index (VARI) for height and exponential or power models using Green Normalized Difference Vegetation Index (GNDVI) for LAI. Conclusion: UAV-derived multispectral VIs showed potential for estimating and spatially modeling ‘Tifton 85’ parameters, especially LAI, under heterogeneous tropical field conditions.

The present study was conducted at the Federal Rural University of Rio de Janeiro, in the years 2024 and 2025, to evaluate the performance of three supervised classifiers - maximum likelihood, random forest, and support vector machine - for spectral discrimination between maize (Zea mays L.) and the weed (Cyperus rotundus), considering different phenological stages and two cropping seasons (main and second crop), using multispectral images acquired by remotely piloted aircraft. Overall accuracy, as well as class metrics of precision, recall, and F1-score, were used to evaluate the performance of the algorithms in classifying four targets: maize, weed, soil, and shadow. The performance of the classifiers improved as the phenological stage progressed. In the first year (second crop), the V8 phenological stage showed the greatest differentiation between the plants, and the highest-performing classifier was maximum likelihood, with an overall accuracy of 89%. For the maize class, precision was 0.88, recall was 0.81, and F1-score was 0.85, and for the weed class, precision was 0.68, recall was 0.89, and F1-score was 0.77. In the second year (main crop), the best differentiation occurred at the V6 stage, and the maximum likelihood classifier again presented the best performance, with an overall accuracy of 88%, precision of 1.00, recall of 0.78, and F1-score of 0.88 for the maize class, and precision of 0.52, recall of 1.00, and F1-score of 0.68 for the weed class. Thus, across the years analyzed, the maximum likelihood classifier performed best.

Imagens de sensores remotos são amplamente utilizadas em aplicações agronômicas e ambientais. No entanto, para uma correta interpretação, exige-se a conversão dos dados em grandezas físicas, o que torna essencial a calibração dos sensores a bordo dos satélites, garantindo confiabilidade das informações. A calibração cruzada, método que compara a resposta de dois sensores expostos a condições semelhantes de radiação incidente, destaca-se por sua economicidade e pela redução das incertezas associadas. Este trabalho analisa a validação da calibração cruzada do sensor MUX do satélite CBERS-4A em relação ao sensor OLI do Landsat 9, adotado como referência. A metodologia para a validação baseou-se na aplicação de coeficientes de calibração previamente determinados ao sensor MUX do CBERS-4A e com base em amostras de áreas homogêneas (água, solo exposto e área pavimentada). O resultado do cálculo das diferenças percentuais de reflectância ficaram abaixo de 3,5%, para os quatro canais dos sensores analisados. Palavras-chave: Coeficientes de calibração; Sensoriamento Remoto; Reflectância planetária ou exoatmosférica.

Este estudo identifica as classes de vegetação, areia e água na praia de Figueira da Foz utilizando imagens da Aeronave Remotamente Pilotada (ARP) Matrice 600 Pro e do satélite Sentinel-2A, com processamento via Orfeo Toolbox no QGIS. A metodologia empregada incluiu o algoritmo Random Forest e técnicas de extração como Haralick, operações morfológicas em escala de cinza e índices radiométricos. As amostras foram divididas para treinamento, teste e validação usando a técnica de hold-out validation. A análise das imagens revelou que a ortofoto obteve uma acurácia de cerca de 94%, superior aos 87% do Sentinel-2A. As métricas de precisão, recall e F1-score foram avaliadas através da matriz de confusão. Os resultados indicam que diferentes técnicas e produtos podem influenciar a escolha do mapeamento e metodologias de classificação, destacando a importância do Sensoriamento Remoto na observação e monitoramento da superfície terrestre.