1. Identity statement | |
Reference Type | Book or Monograph (Book) |
Site | mtc-m12.sid.inpe.br |
Holder Code | isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S |
Repository | bol.com.br/mirian_cris/2003/01.17.09.38 |
Last Update | 2016:03.08.14.41.52 (UTC) jeferson |
Metadata Repository | bol.com.br/mirian_cris/2003/01.17.09.38.53 |
Metadata Last Update | 2019:02.04.13.03.15 (UTC) administrator |
Secondary Key | INPE-8971-NTC/349 |
Label | 9987 |
Citation Key | AdamiPinhMore:2002:ApDiAl |
Title | Aplicação de diferentes algoritmos para a classificação de imagens ETM+/Landsat-7 no mapeamento agrícola |
Year | 2002 |
Secondary Date | 20020916 |
Access Date | 2024, May 15 |
Secondary Type | NTC |
Number of Pages | 41 |
Number of Files | 5 |
Size | 872 KiB |
|
2. Context | |
Author | 1 Adami, Marcos 2 Pinheiro, Eduardo da Silva 3 Moreira, Maurício Alves |
Resume Identifier | 1 2 3 8JMKD3MGP5W/3C9JHT4 |
Group | 1 DSR-INPE-MCT-BR 2 DSR-INPE-MCT-BR |
Publisher | INPE |
City | São José dos Campos |
History (UTC) | 2005-07-20 16:11:52 :: banon -> administrator :: 2016-03-08 14:40:03 :: administrator -> jeferson :: 2002 2016-03-08 14:41:52 :: jeferson -> administrator :: 2002 2019-02-04 13:03:15 :: administrator -> simone :: 2002 |
|
3. Content and structure | |
Is the master or a copy? | is the master |
Content Stage | completed |
Transferable | 1 |
Keywords | AGRONOMIA Leópolis (PA)Rancho Alegre (PA)Sertaneja (PA) classificação de imagens algoritmos agricultura NDVI índice de vegetação da diferença normalizada modelo mistura mapeador temático de realce (Landsat) mapeador temático (Landsat) mixtures models Landsat 7 agriculture image classification |
Abstract | Atualmente, o grande volume de dados coletados por satélites de recursos naturais e o desenvolvimento da informática, têm estimulado o aparecimento de muitas técnicas para o processamento de imagens digitais. 0 presente trabalho, procura analisar o desempenho de alguns algoritmos para a classificação de imagens do sensor ETM+ /Landsat-7, visando mapear o uso e cobertura do solo em três municípios do Estado do Paraná (Leópolis, Rancho Alegre, Sertaneja). Este mapeamento buscou discernir as classes: agricultura, solo exposto, pastagem, mata e corpos d'água. Além das imagens originais, também foram testadas as transformações NDVI, Principais Componentes e Modelo Linear de Mistura Espectral para verificar se ocasionariam melhoras nas classificações. Os classificadores utilizados foram K-médias, Isoseg, Máxima Verossimilhança, Distância de Mahalanobis e Distância Bhattacharyya. Para avaliar a exatidão de mapeamento utilizou-se matriz de confusão e o coeficiente Kappa. Foi considerada como verdade terrestre a combinação de duas classificações visuais, padronizadas por um algoritmo LEGAL/SPRING. Concluiu-se que os melhores desempenhos de classificação foram obtidos pelo classificador Isoseg e Bhattacharyya, quando aplicados nos dados originais do ETM+ das bandas 3, 4 e 5. ABSTRACT: Nowadays, the huge amount of data acquired by earth observation satellites and the development of computer technology have stimulated the appearance of several digital image processing techniques. In the present work it was analyzed the performance of five classification algorithms on Landsat-7 images in order to identify and map soil use and coverage in three municipalities (Leópolis, Rancho Alegre, Sertaneja)in the State of Parand, Brazil. For the mapping phase it was selected me following classes: agricultural land, bare soil, grassland, woodland and water bodies. Original ETM/Landsat-7 images were transformed into NDVI, Principal Components Method and Linear Spectral Mixture Model in order to verify influence of each algorithm on the classification results. The following classification algorithms were used: K-Medias, Isoseg, Maximum Likelihood, Mahalanobis Distance and Bhattacharyya Distance. Ground reference data were obtained from visual classification performed by two photointerpreters. These visual classification were, Aerwards, standardized by LEGAl/SPRING algoritbm in order to generate the error matrix and Kappa coefficient. The best classification performances were obtained through Isoseg and Bhattacharyya classifiers when they were directly applied to ETM+ original data bands 3, 4 and 5. |
Area | SRE |
Arrangement | urlib.net > BDMCI > Fonds > Produção anterior à 2021 > DIDSR > Aplicação de diferentes... |
doc Directory Content | access |
source Directory Content | there are no files |
agreement Directory Content | there are no files |
|
4. Conditions of access and use | |
data URL | http://urlib.net/ibi/bol.com.br/mirian_cris/2003/01.17.09.38 |
zipped data URL | http://urlib.net/zip/bol.com.br/mirian_cris/2003/01.17.09.38 |
Language | pt |
Target File | publicacao.pdf |
User Group | administrator jeferson |
Visibility | shown |
Update Permission | not transferred |
|
5. Allied materials | |
Next Higher Units | 8JMKD3MGPCW/3ER446E |
Citing Item List | sid.inpe.br/mtc-m21/2012/07.13.14.56.26 2 |
Dissemination | NTRSNASA, BNDEPOSITOLEGAL. |
Host Collection | sid.inpe.br/banon/2001/04.06.10.52 |
|
6. Notes | |
Empty Fields | affiliation archivingpolicy archivist callnumber contenttype copyholder copyright creatorhistory descriptionlevel doi e-mailaddress edition editor electronicmailaddress format identifier isbn issn lineage mark mirrorrepository nextedition notes numberofvolumes orcid parameterlist parentrepositories previousedition previouslowerunit progress project readergroup readpermission rightsholder schedulinginformation secondarymark serieseditor seriestitle session shorttitle sponsor subject tertiarymark tertiarytype translator url versiontype volume |
|
7. Description control | |
e-Mail (login) | simone |
update | |
|