Mapping of the spatial distribution and temporal variation of rupestrian vegetation in Serra do Cipó (MG) through images obtained by remotely piloted aircraft

Carregando...
Imagem de Miniatura

Data

2019-10-16

Autores

Medeiros, Thais Pereira de

Título da Revista

ISSN da Revista

Título de Volume

Editor

Universidade Estadual Paulista (Unesp)

Resumo

The Serra do Cipó, located in Cadeia do Espinhaço, north of Belo Horizonte (MG), inside of the domain of the Cerrado and with predominance of phytophysiognomy of rupestrian grassland, is characterized by a high richness and endemicity of herbaceous species. Cadeia do Espinhaço is the center of diversity of various group of plants, including more than 4000 species and, the region of Serra do Cipó, covering approximately 200 kilometers inside of the Espinhaço, shelter more than a third of this biodiversity. These regions are currently under great threat, mainly by tourism activity and mining. Therefore, is extremely important the development of new methods and approaches that allow the effective monitoring of these environments, and so making possible to better understand its operation and vulnerabilities. The Unmanned Aerial Vehicles (UAVs), also known as Remotely Piloted Aircraft (RPAs), are equipments that contribute for the studies and monitoring of both the environmental and ecologic vulnerabilities of such environments. The OBIA (Object Based Image Analysis) classification method has been used successfully for the analysis of high resolution images. However, they still present many difficulties for the automatic classification of surface elements, due to the extremely high spatial resolution of their sensors. The use of these technologies for the monitoring of heterogeneous vegetation in actually is still challenging due to the great structural and spatial heterogeneity existing in these environments. In this context, the goal of the research was answer the following questions: What are the better combinations of parameters for proposing a high quality of ultrahigh spatial resolution classification? Is the OBIA an efficient method for classifying an heterogeneous ecosystem, such as Serra do Cipó? Does the spatial resolution influence species classification? What is the optimal spatial resolution for classifying an image with the ultrahigh spatial resolution?
A Serra do Cipó, localizada na Cadeia do Espinhaço, ao norte de Belo Horizonte (MG), dentro do domínio do Cerrado e com predominância de fisionomia de campos rupestres, é caracterizada por uma alta riqueza e endemicidade de espécies herbáceas. A Cadeia do Espinhaço é o centro de diversidade de vários grupos de plantas, incluindo mais de 4000 espécies e, a região da Serra do Cipó, abrangendo aproximadamente 200 km² dentro do Espinhaço, abriga mais que um terço dessa biodiversidade. Estas regiões encontram-se atualmente sob grande ameaça, principalmente pela atividade turística e pela mineração. Portanto, é de extrema importância o desenvolvimento de novos métodos e abordagens que permitam o monitoramento eficaz destes ambientes, e assim possibilitem uma melhor compreensão de seu funcionamento e vulnerabilidades. Os Veículos Aéreos Não Tripulados (VANTs), também conhecidos como Aeronaves Remotamente Pilotadas (ARP) são equipamentos que contribuem para os estudos e monitoramento das vulnerabilidades ambientais e ecológicas desses ambientes. O método OBIA (Object Based Image Analysis) de classificação tem sido utilizado com sucesso para a realização de análises a partir das imagens de alta resolução. Entretanto, ainda existem dificuldades para a classificação automática de elementos da superfície, em imagens obtidas por VANTs, devido à altíssima resolução espacial de seus sensores. O uso dessas tecnologias para o monitoramento de vegetações heterogêneas na atualidade ainda é desafiador por conta da grande heterogeneidade estrutural e espacial existente nestes ambientes. Neste contexto, o objetivo do trabalho foi responder as seguintes perguntas: Qual a melhor combinação de parâmetros para obtenção de uma alta qualidade na classificação de imagem com altíssima resolução espacial? O método OBIA é eficiente para classificar ecossistemas heterogêneos, como por exemplo, a Serra do Cipó? A resolução especial das imagens apresenta influencia no processo de classificação? Qual é a resolução especial ideal para classificar uma imagem com altíssima resolução espacial?

Descrição

Palavras-chave

UAS, Rupestrian grassland, OBIA method, Remote sensing, Unmanned Aerial System, Biodiversity, UAS classification method, Campo rupestre, Método OBIA, VANT

Como citar