Avaliação da acurácia de produtos cartográficos adquiridos a partir de fotogrametria com veículo aéreo não tripulado: estudo de caso na fazenda belém e soledade no município de pão de açúcar

Abstract

Photogrammetry using Unmanned Aerial Vehicles (UAVs) has become an important tool for topographic surveying, enabling the production of high-resolution cartographic products with greater operational efficiency and reduced acquisition costs. However, the use of these products in technical applications requires the assessment of their positional quality. This study aimed to evaluate the positional quality of cartographic products generated from a UAV photogrammetric survey conducted at Belém and Soledade Farm, located in the municipality of Pão de Açúcar, Alagoas, Brazil. The evaluation was performed by comparing the photogrammetric products with GNSS RTK survey data and by applying the Brazilian Cartographic Accuracy Standard for Digital Cartographic Products (PEC-PCD) using the GeoPEC software. The survey covered an area of approximately 151 hectares, resulting in the generation of an orthomosaic, Digital Surface Model (DSM), Digital Terrain Model (DTM), dense point cloud, and contour lines. The results showed that the generated products met the requirements for PEC-PCD Class A at the 1:1,000 scale, with a planimetric Root Mean Square Error (RMSE) of approximately 0.96 cm. The Shapiro–Wilk test confirmed the normal distribution of positional errors, whereas Student's t-test detected the presence of systematic bias, indicating high precision but reduced positional accuracy due to systematic errors. The main limitation of the study was the limited number of check points used in statistical validation. It is concluded that UAV photogrammetry combined with GNSS RTK is an effective approach for generating cartographic products for topographic surveying, although the detected systematic bias highlights the need for improvements in the validation procedure in future studies.