Uso de aeronave remotamente pilotada (RPA) para avaliação da soja sob diferentes fungicidas

Abstract

The cultivation of soybean has high relevance in Brazilian agribusiness, being directly impacted by phytosanitary factors that require the use of fungicides for disease control. However, the growing search for sustainable practices has driven the adoption of bioinputs and technologies that enable more efficient and less subjective assessments. In this context, the present study aimed to relate vegetation indices obtained using a remotely piloted aircraft and an imaging sensor operating in the visible spectrum in soybean cultivation under different fungicide protocols. The evaluated experiment was conducted by Pasqua (2025) in the experimental area of the Federal University of Technology – Paraná, Dois Vizinhos Campus, with an experimental design composed of thirteen treatments and four replications, including combinations of conventional fungicides and biofungicides. Aerial images were acquired through periodic flights, following field assessments, using a remotely piloted aircraft (RPA) equipped with an imaging sensor operating in the visible spectrum, and were subsequently processed for orthomosaic generation and vegetation index calculation, namely: Green Leaf Index (GLI), Visible Atmospherically Resistant Index (VARI), Normalized Green-Red Difference Index (NGRDI), and Triangular Greenness Index (TGI). From these data, quantitative metrics were extracted, and temporal dynamics of vegetation cover were analyzed, as well as their correlation with disease severity and yield. The results indicated that the evaluated vegetation indices exhibited distinct behaviors, with GLI being the most effective in identifying vegetation cover over time. Regarding the correlation with disease severity, low-magnitude associations were observed, possibly due to the reduced severity intensity recorded in the experiment. On the other hand, significant differences among treatments were identified concerning the maintenance of vegetation cover, with those receiving chemical fungicide applications showing greater persistence throughout the crop cycle. It is concluded that the use of RPA associated with digital image processing constitutes an efficient tool to complement the phytosanitary assessment of soybean.