UTILIZAÇÃO DE DRONE DE BAIXO CUSTO PARA DETERMINAÇÃO DE FALHAS DE PLANTIO EM CANA-DE-AÇÚCAR

Abstract

One of the factors that impact sugarcane production is planting failures. The study aimed to identify and map sugarcane planting failures through images captured by a low-cost drone equipped with an RGB camera, and to determine the best plant height to identify these failures. The experiment was conducted in the municipality of Ceres - GO. The experimental area was 4.75 ha with cv. SP 80-1816, planted with a spacing of 1.5 m between rows. The images were collected by the DJI Mavic Mini drone and measurements of failures and plant height were taken at 30, 60 and 90 days after planting (DAP). Four failure lengths were collected: 0.5 to 1.0 m; 1.0 to 1.5 m; 1.5 to 2.0 m, and above 2.0 m, in 10 replicates, manually in the area, for comparison with the aerial images. The orthomosaic was processed and generated. The QGIS software program was used to identify flaws in the images and compare the data collected in the field with a tape measure. The flight height and crop development stage in mapping flaws in sugarcane fields interact with the accuracy of the flaw estimate compared to field observations. The results indicated that the quality of the mapping is influenced by the crop development and the appropriate time to capture the images. Mapping flaws in plants up to 0.5 m in height, which corresponds to 60 DAP, reduces errors. The accuracy decreases with smaller flaws when exceeding 60 DAP, which is a limitation of the study in the mapping of 0.5 m flaws.