CLASSIFICAÇÃO DE ÁREAS IRRIGADAS POR PIVÔS CENTRAIS UTILIZANDO COMO BASE A SEGMENTAÇÃO
DOI:
https://doi.org/10.15809/irriga.2011v16n2p145Abstract
1 RESUMO
O uso de técnicas de análise de dados de sensoriamento remoto, baseadas na segmentação de objetos homogêneos, abre novas perspectivas para a estimativa de áreas irrigadas por pivôs centrais, pois elas possuem formas bem definidas. O objetivo deste trabalho foi comparar e avaliar mapas de áreas irrigadas por pivôs centrais gerados a partir de dois métodos: interpretação visual e classificação por regiões. As etapas do trabalho envolveram o georreferenciamento e correção geométrica da imagem de imagens de satélite CCD/CBERS-2B, a interpretação visual das áreas irrigadas, a segmentação e a classificação por região para extração das feições. As áreas irrigadas apresentaram-se em diferentes cores e tonalidades, representando culturas agrícolas em diferentes estádios e solo exposto. Os valores totais das áreas irrigadas foram de 631,38 ha, para o método manual e de 669,12 ha utilizando-se o método baseado na segmentação. Conclui-se que houve uma concordância excelente entre as áreas delimitadas manualmente e as áreas extraídas com base na segmentação, com índice kappa de 94%. O método baseado na segmentação não individualizou os pivôs centrais, mas permitiu identificar e estimar o total das áreas irrigadas. A extração das áreas irrigadas por pivôs centrais, utilizando o processo de segmentação, dependeu da distribuição espacial dos mesmos.
UNITERMOS: CCD/CBERS-2B, processamento de imagens, agricultura irrigada.
FERREIRA, E.; DANTAS, A. A. A.; TOLEDO, J. H.
CLASSIFICATION OF IRRIGATED AREAS BY CENTER PIVOT BASED ON THE SEGMENTATION TECHNIQUE.
2 ABSTRACT
The use of techniques of analysis of remote sensing data based on homogeneous regions segmentation opens new perspectives for center pivots irrigated areas estimation, because of their well-defined shapes. The goal of this work was to compare and assess maps of irrigated areas by center pivots using two interpretation methods: visual interpretation and a classification based on segmentation techniques. The thematic maps were generated from CCD/CBERS-2B satellite images. The steps of the work involved the geo-reference and the image geometric correction, the visual interpretation of irrigated areas, segmentation, and supervised classification for extraction of features. The irrigated areas were presented in different colors and shades, representing agricultural crops in different stages and soil exposed. The total values of irrigated areas were 631.38 ha, for the manual method and 669.12 ha for the segmentation method. It was concluded that there was an excellent agreement between the areas that were delineated manually and the areas extracted with the method based on segmentation, with kappa index of 94%. The method based on segmentation did not separate the center pivots, but allowed to identify and estimate the total irrigated areas. The extraction of irrigated areas by center pivots, using the process of segmentation, depended on the spatial distribution of the objects.
KEYWORDS: CCD/CBERS-2B, image processing, irrigated agriculture.
Downloads
Published
How to Cite
Issue
Section
License
This magazine provides public access to all its content, following the principle that free access to research generates greater global knowledge exchange. Such access is associated with a growing reading and citation of an author's work. The copyright of articles published in Revista Irriga is the property of the authors, with first publication rights for the journal. Because they appear in this publicly accessible magazine, articles are free to use, for their own purposes, for educational and non-commercial purposes. Further details can be obtained at http://creativecommons.org/licenses/by-nc-nd/4.0