EXTRAÇÃO DE INFORMAÇÃO DA CENA URBANA COMBINANDO IMAGENS MULTIESPECTRAIS DE ALTA RESOLUÇÃO E IMAGENS RADAR
DOI:
https://doi.org/10.5380/bcg.v17i4.25415Abstract
Os sistemas de imageamento por microondas tem sido objeto de
estudo de pesquisas aplicadas a diversos fenômenos. O sistema RADAR
(RAdio Detection and Ranging) é um sistema ativo, que gera imagens
utilizando energia eletromagnética na faixa de micro-ondas. As imagens geradas
por sensores ativos como o RADARSAT-2, tem características diferentes das
imagens ópticas. Apesar da alta resolução disponível em imagens ópticas como
a GeoEye, a informação contida nestas imagens é insuficiente para a
classificação dos elementos de uma cena urbana. Uma alternativa para a
obtenção de informações confiáveis sobre os elementos da cena é a combinação
das informações de imagens óptica e radar. Um método para a classificação de
dados de fontes diferentes não pode ser baseado apenas em cálculos estatísticos.
A utilização de árvores de decisão contorna esse problema e permite que se
proceda a classificação supervisionada através de um algoritmo com base em
regras. Assim será possível integrar as informações disponíveis e classificar a
imagem segundo as classes de interesse.
Downloads
Published
How to Cite
Issue
Section
License
Submission of an original manuscript to the Journal will be taken to mean that it represents original work not previously published, that is not being considered elsewhere for publication.
The BCG allows the author(s) to hold the copyright without restrictions and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
The BCG also allows the authors to retain publishing rights without restrictions.
