Aeronave remotamente pilotada (RPA): Análise bibliométrica em estudos agrícolas e estimativa da contagem de plantas de milho (Zea mays) através de imagens aéreas
Arquivos
Data
2021-09-02
Autores
Otani, Tiago Makoto
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
O uso do sensoriamento remoto (SR) na agricultura vem sendo amplamente utilizada principalmente em extensas áreas na inferência sobre os processos biológicos das plantas. Neste contexto, para um melhor entendimento sobre a evolução dos parâmetros de publicações sobre as plataformas de SR de baixa cobertura em área, como as Aeronaves Remotamente Pilotadas (RPA), primeiramente, fez-se o estudo de revisão através da análise bibliométrica. O estudo foi feito com dados da plataforma Scopus (Elsevier), considerando os últimos 26 anos, quanto ao número de publicações, tipos de publicações, revistas, países, instituições de pesquisa, e palavras-chave. A pesquisa com base nas RPAs demonstrou ser um campo de estudo amplamente fragmentada. Os principais tópicos estão intimamente relacionados com o monitoramento das lavouras assim como no uso da agricultura de precisão, focados principalmente em estudos para a real aplicação e com alta acurácia, como os dados obtidos a partir de laser e radar para análise de estrutura de dossel e mapeamento de vegetação. No segundo capítulo, a aplicação com o uso de RPA foi apresentada e avaliada através de um fluxograma combinando as RPAs e programa de processamento de código aberto e livre na contagem de plantas de milho em estágio vegetativo inicial em condições de campo comercial na de grãos. O fluxograma implementado demonstrou ser capaz de detectar e contar plantas de milho. A avaliação da performance deste fluxograma é dependente da resolução espacial das imagens adquiridas, avaliadas como a principal limitação neste trabalho. Limitações e pesquisas futuras são apresentadas no decorrer do artigo.
The use of remote sensing (SR) in agriculture has been widely spread mainly in extensive areas in inferring plants biological processes. In this context, for a better understanding on the growth of publications on SR platforms with low area coverage, such as Remotely Piloted Aircraft (RPA), firstly, a review study was carried out through bibliometric analysis. The study was carried out with data from the Scopus platform (Elsevier), considering the last 26 years, regarding the number of publications, types of publications, journals, countries, research institutions, and keywords. RPA-based research has proven to be a largely fragmented field of study. The main topics are closely related to crop monitoring as well as the use of precision agriculture, mainly focused on studies for real application and with high accuracy, such as laser and radar data for canopy structure analysis and vegetation mapping. In the second chapter, the application with the use of RPA was presented and evaluated through a workflow combining the RPAs and an open-source processing program in counting early vegetative-stage maize plants under commercial grain field conditions. The implemented workflow proved to be able to detect and count corn plants. The evaluation of the performance of this workflow depends on the spatial resolution of the acquired images, evaluated as the main limitation in this work. Limitations and future research are presented throughout the article.
The use of remote sensing (SR) in agriculture has been widely spread mainly in extensive areas in inferring plants biological processes. In this context, for a better understanding on the growth of publications on SR platforms with low area coverage, such as Remotely Piloted Aircraft (RPA), firstly, a review study was carried out through bibliometric analysis. The study was carried out with data from the Scopus platform (Elsevier), considering the last 26 years, regarding the number of publications, types of publications, journals, countries, research institutions, and keywords. RPA-based research has proven to be a largely fragmented field of study. The main topics are closely related to crop monitoring as well as the use of precision agriculture, mainly focused on studies for real application and with high accuracy, such as laser and radar data for canopy structure analysis and vegetation mapping. In the second chapter, the application with the use of RPA was presented and evaluated through a workflow combining the RPAs and an open-source processing program in counting early vegetative-stage maize plants under commercial grain field conditions. The implemented workflow proved to be able to detect and count corn plants. The evaluation of the performance of this workflow depends on the spatial resolution of the acquired images, evaluated as the main limitation in this work. Limitations and future research are presented throughout the article.
Descrição
Palavras-chave
aeronave remotamente pilotada, milho, agricultura de precisão, remotely piloted aircraft, maize, precision agriculture