Análise de falha de brotação da cana-de-açúcar através de metodologia manual e geoprocessamento em um sistema de cultivo mecanizado

• Autor(a) principal: MELO, Camila Gomes Bezerra de
• Data de Publicação: 2022
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFRPE
• Texto Completo: http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/9072
• topic: Agricultura de precisão; Cana-de-açúcar; Plantio; Falha de brotação; Geoprocessamento; Colheita mecanizada; CIENCIAS AGRARIAS::ENGENHARIA AGRICOLA

The sugarcane crop cycle can last an average of five years before replanting, with each harvest being made annually or one year and a half. However, during the production cycles, it is common for sprouting row gaps to appear, which intensify after successive harvests. Sugarcane row gaps are empty spaces without stalks that occur in the sugarcane rows and are thus associated with decreased productivity, and their measurement/quantification is commonly performed on site manually. Thus, this work aimed to propose an alternative methodology for the measurement of planting and regrowth gaps in sugarcane crops from images acquired with VANT, using geoprocessing data, and to evaluate the quality of the measurement of the gap compared with the manual measurement performed in the field. In an experimental area of one hectare, under mechanized harvesting, an experiment was conducted to analyze the behavior of the sugarcane planting and regrowth gaps. The reference methodology consists in measuring the continuous distances without sprouting between two sugarcane stalks along a planting line, considering as faults those distances larger than 0.50 m. The measurement is performed 90 days after planting when the crop is already established. While in the measurement using an unmanned aerial vehicle, aerial images of RGB type (Red, Green, Blue bands) with 12 megapixels (4000 × 3000 pixels) were obtained and processed to generate the respective ortomosaics of the final images. After this, the images were processed with the processing algorithm in QGIS software, and the results were compared with the values obtained manually after planting and harvesting. The number of sprouting faults measured through the UAV image was higher with lengths between 0.5-1.0 m, however in the field it was lower. In the class of faults between 1-1.5 m, the result is inverted. In the other length classes analyzed the number of faults did not showed a representative difference