Desempenho de um vant na determinação de índices de vegetação da cultura de crambe

• Autor(a) principal: Felipetto, Henrique dos Santos
• Data de Publicação: 2016
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações do UNIOESTE
• Texto Completo: http://tede.unioeste.br:8080/tede/handle/tede/2729
• topic: Sensoriamento Remoto; VANT; Índices de Vegetação; Agricultura de Precisão; Remote Sensing; UAV; Vegetation Index; Precision Agriculture; CNPQ::CIENCIAS AGRARIAS::ENGENHARIA AGRICOLA

The use of orbital remote sensing (RS) techniques has come out as a tool for decision making in precision agriculture and has been intensified through time. However, sometimes the high cost of this technology, as well as low temporal and spatial resolution are unfeasible with such technique mainly in small areas. Thus, this study aimed at analyzing thoroughtly the use of an unmanned aerial vehicle UAV, at low cost of development and establishment to determine the green indices of crambe crop in the western region of Paraná. The applied methodology developed up from the purchase of equipment until the UAV setting to determine RNIR NDVI and SAVI green indices, based on the images produced by the equipment. It was carried out a comparative with terrestrial to validate data with an active GreenSeeker sensor and a passive Spectroradiometer. As a result, three indices of vegetation for the three sensors were created and statistically compared using Spearman's correlation coefficient at 5% significance level. The studied equipment showed some satisfactory performance and there was no anomaly, since it followed the National Civil Aviation Agency (ANAC) safety standards. According to the economic viability, it was also positive since the setting and development costs did not exceed 10% value of the most part of the equipment currently sold in Brazil. According to the index values generated by sensors entrained in the UAV, it was possible to determine each stage of cranberry plant development, totaling seven flights during it scropcycle. When the passive sensor coupled to the UAV was compared to active field sensors GreenSeeker and Spectroradiometer, the UAV showed a good performance to determine RNIR, NDVI and SAVI indices. Consequently, there was a significant correlation at 5%level only at haying period, which corresponded to blooming and graining start of crambe, for all indices