Tratamento de imagens orbitais e suborbitais para caracterização ambiental da cabeceira do Rio São Lourenço-MT
Carregando...
Data
2012-12-14
Autores
Cândido, Anny Keli Aparecida Alves
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal de Mato Grosso
Resumo
The thesis has been divided into two chapters. In the first chapter aimed to identify potential
environmental changes that occurred over a period of 26 years. For this we used satellite
imagery scenes of Landsat TM 5 of 1985, 1997 and 2011, for procedures supervised
classification by region, with the maximum likelihood algorithm. From these images, we
obtained maps of NDVI from the subtraction of wavelengths near infrared and red, divided by
the sum of the ratio of wavelengths of near infrared and red spectrum. In mapping land use, it
was found that over 80% of the watershed area is allocated to agriculture. The reduced class
savannah and forest class increased over the years analyzed. The NDVI showed areas that
were appropriately with little plant biomass, differing areas with dense vegetation cover.
Regarding the detection of degraded areas in the watershed, there was a demo unit of
ecological restoration (UDRE), established in 2010, in terms of spectral response, obtained
via NDVI, is similar to the class agriculture. It is believed that this area of permanent
preservation degraded in the recovery phase, tends to become similar areas with dense
vegetation, over the next few years. The second is a chapter emphasized the treatment of
aerial photographs of high spatial resolution, obtained by unmanned aerial vehicle (UAV),
aiming at environmental characterization of the watershed is located where the headwaters of
the St. Lawrence River. We tested several routines supervised classification and unsupervised
pixel by pixel and by region, using two Geographic Information Systems. The supervised
classification by region segmentation 20 and 200 of area similarity, showed statistical
similarity with the conventional procedure performed by photointerpretation. The techniques
presented in this research will be useful for jobs that involve monitoring of degraded areas,
bringing effective contribution to setting the methodological route plan reclamation prevailing
in the state of Mato Grosso.
Descrição
Palavras-chave
Citação
CÂNDIDO, Anny Keli Aparecida Alves. Tratamento de imagens orbitais e suborbitais para caracterização ambiental da cabeceira do Rio São Lourenço-MT. 2012. 94 f. Dissertação (Mestrado em Engenharia Agrícola) - Universidade Federal de Mato Grosso, Instituto de Ciências Agrárias e Tecnológicas, Rondonópolis, 2012.