1. Identity statement | |
Reference Type | Conference Paper (Conference Proceedings) |
Site | marte.sid.inpe.br |
Holder Code | isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S |
Repository | dpi.inpe.br/sbsr@80/2008/11.17.13.56.38 |
Last Update | 2009:03.30.21.42.52 (UTC) administrator |
Metadata Repository | dpi.inpe.br/sbsr@80/2008/11.17.13.56.39 |
Metadata Last Update | 2022:07.07.03.53.18 (UTC) administrator |
Secondary Key | INPE-15966-PRE/10575 |
ISBN | 978-85-17-00044-7 |
Citation Key | OliveiraSouBraSilTei:2009:ApMaKo |
Title | Aplicação de Mapas de Kohonen em imagem de satélite do semi-árido e comparação com o método da máxima verossimilhança |
Format | DVD, On-line. |
Year | 2009 |
Access Date | 2024, May 18 |
Secondary Type | PRE CN |
Number of Files | 1 |
Size | 1002 KiB |
|
2. Context | |
Author | 1 Oliveira, Davi Nunes de 2 Sousa, Beatriz Fernandes Simplicio 3 Braga, Arthur Plínio de Souza 4 Silva, Francisco de Assis Tavares Ferreira da 5 Teixeira, Adunias dos Santos |
Resume Identifier | 1 2 3 4 8JMKD3MGP5W/3C9JH4L |
Group | 1 2 3 4 CRN-CCR-INPE-MCT-BR |
Affiliation | 1 Universidade Federal do Ceará 2 Universidade Federal do Ceará 3 Universidade Federal do Ceará 4 Instituto Nacional de Pesquisas Espaciais 5 Universidade Federal do Ceará |
Author e-Mail Address | 1 davi@dee.ufc.br 2 beatrizsimplicio@yahoo.com.br 3 arthurp@ufc.br 4 tavares@roen.inpe.br 5 adunias@ufc.br |
Editor | Epiphanio, José Carlos Neves Galvão, Lênio Soares |
e-Mail Address | lise@dpi.inpe.br |
Conference Name | Simpósio Brasileiro de Sensoriamento Remoto, 14 (SBSR) |
Conference Location | Natal |
Date | 25-30 abr. 2009 |
Publisher | Instituto Nacional de Pesquisas Espaciais (INPE) |
Publisher City | São José dos Campos |
Pages | 7877-7884 |
Book Title | Anais |
Tertiary Type | Artigo |
Organization | Instituto Nacional de Pesquisas Espaciais (INPE) |
History (UTC) | 2008-11-17 14:23:04 :: beatrizsimplicio@yahoo.com.br -> sbsr :: 2008-12-15 19:35:14 :: sbsr -> lise@dpi.inpe.br :: 2009-01-15 17:26:09 :: lise@dpi.inpe.br -> sbsr :: 2009-02-19 20:15:51 :: sbsr -> administrator :: 2009-05-28 12:59:56 :: administrator -> lise@dpi.inpe.br :: 2009-06-16 13:14:54 :: lise@dpi.inpe.br -> marciana :: 2009-06-22 13:55:17 :: marciana -> erich@sid.inpe.br :: 2009-06-26 14:11:36 :: erich@sid.inpe.br -> administrator :: 2009-08-05 01:55:44 :: administrator -> erich@sid.inpe.br :: 2010-05-14 02:40:34 :: erich@sid.inpe.br -> marciana :: 2011-02-16 14:17:05 :: marciana -> administrator :: 2009 2022-07-07 03:53:18 :: administrator -> :: 2009 |
|
3. Content and structure | |
Is the master or a copy? | is the master |
Content Stage | completed |
Transferable | 1 |
Keywords | remote sensing image processing artificial neural networks confusion matrix sensoriamento |
Abstract | Artificial Neural Networks represent a potential approach for classification of remote sensing images. Different from the traditional multi-layer neural networks, the Kohonen self-organizing map is a two-layer network that creates class representation by self organizing the connection weights from the input patterns to the output layer. This paper aims to compare the performance of Kohonen self-organizing map with classification method of maximum likelihood to classify the caatinga biome. The test of the algorithm was carried out by classifying the scene of satellite Landsat Thematic Mapper (bands 2,3,4). This scene contains the city of Iguatu, Ceara, Brazil and the study area was a piece of this city. Iguatu is a semi-arid region that contains an important, fragile and depleted biome: caatinga. A quantitative validation was carried out by comparison between 92 GPS points collected, and used as a ground truth, and the image obtained of each classification proposed. Using these points it was generated a confusion matrix to each classification. Also, the Kappa coefficient and overall accuracy have been calculated. The experimental results show that the self-organizing map presented an equal performance to the Maximum Likelihood algorithm with values of Kappa coefficient of 0,69. In this case, classifying caatinga biome through the analyzes of the specific accuracy, the self-organizing map has a better performance than the maximum likelihood method. |
Area | SRE |
Type | Técnicas de Classificação e Mineração de Dados |
Arrangement | urlib.net > BDMCI > Fonds > Produção anterior à 2021 > CRCRN > Aplicação de Mapas... |
doc Directory Content | access |
source Directory Content | artigo_ko_Beatriz.pdf | 17/11/2008 12:23 | 456.9 KiB | |
agreement Directory Content | there are no files |
|
4. Conditions of access and use | |
data URL | http://urlib.net/ibi/dpi.inpe.br/sbsr@80/2008/11.17.13.56.38 |
zipped data URL | http://urlib.net/zip/dpi.inpe.br/sbsr@80/2008/11.17.13.56.38 |
Language | pt |
Target File | 7877-7884.pdf |
User Group | administrator beatrizsimplicio@yahoo.com.br lise@dpi.inpe.br marciana erich@sid.inpe.br administrator |
Reader Group | administrator beatrizsimplicio@yahoo.com.br |
Visibility | shown |
|
5. Allied materials | |
Mirror Repository | dpi.inpe.br/marte1@80/2009/05.26.17.02 |
Next Higher Units | 8JMKD3MGPCW/3EUAPES |
Citing Item List | sid.inpe.br/mtc-m21/2012/07.13.14.46.21 1 |
Host Collection | dpi.inpe.br/netuno@1905/2006/07.17.20.18 dpi.inpe.br/banon/2003/12.10.19.30 |
|
6. Notes | |
Sponsor | CNPq |
Mark | 1 |
Empty Fields | archivingpolicy archivist callnumber contenttype copyholder copyright creatorhistory descriptionlevel dissemination documentstage doi edition identifier issn label lineage nextedition notes numberofvolumes orcid parameterlist parentrepositories previousedition previouslowerunit progress project readpermission rightsholder schedulinginformation secondarydate secondarymark serieseditor shorttitle tertiarymark url versiontype volume |
|