Classificação supervisionada de solos por redes neurais artificiais na Serra do Cipó - MG
Ano de defesa: | 2009 |
---|---|
Autor(a) principal: | |
Orientador(a): | |
Banca de defesa: | , , |
Tipo de documento: | Dissertação |
Tipo de acesso: | Acesso aberto |
Idioma: | por |
Instituição de defesa: |
Universidade Federal de Viçosa
|
Programa de Pós-Graduação: |
Mestrado em Solos e Nutrição de Plantas
|
Departamento: |
Fertilidade do solo e nutrição de plantas; Gênese, Morfologia e Classificação, Mineralogia, Química,
|
País: |
BR
|
Palavras-chave em Português: | |
Palavras-chave em Inglês: | |
Área do conhecimento CNPq: | |
Link de acesso: | http://locus.ufv.br/handle/123456789/5456 |
Resumo: | The supervised classification of soils, especially in recent decades, is being carried out using mathematical and statistical models, amongst which the model of neural networks, stand out by greater accuracy of mapping comparing to classical models, such as Maximum Likelihood (MaxVer), helping the conventional method of mapping. Neural Networks model has been performed mostly for soil properties, with little application for soil classes. This work aimed to undertake the classification of soil using neural networks and MaxVer for an area located in Serra do Cipó, in the State of Minas Gerais. The map units was defined based on information from 55 soil profiles classified accord to the Brazilian System of Soils Classification to the fourth categorical level, whereas compound units were done with basis on the similarity of soil properties and the characteristics of physics environment. The discriminates variables used included: six scenes of Landsat satellite image sensor ETM+, four indexes derived from this image (Clay minerals, Ferrous minerals, Iron oxide and NDVI), Digital Elevation Model and derived attributes: altitude, slope, compound topographic index, aspect, solar radiation, curvature and elevation amplitude, in addiction to geological and soil maps. Several variables combinations were tested in both classifiers, selecting those that best contribute to classify the soil with high accuracy on the two supervised mapping approaches. The classification by neural networks was performed using the Stuttgart Neural Network Simulator and the backpropagation algorithm, the framework and classification parameters were selected by training and statistical tests. The results obtained with both classifiers, neural networks and MaxVer were compared using ground data as reference. The same set of reference points was used to validate the soil map obtained by the conventional method of mapping. Maps obtained by the two classifiers using the group of variables that provided the best performance to the classification showed a good accuracy index, with no statistical difference in overall accuracy of the maps. The map generated by MaxVer showed a kappa index of 0.58, while the map from neural network showed an index of 0.60. Although the accuracy of the two maps was statistically similar, the classifiers efficiency in individual discrimination of soil units differed significantly, with two units being best classified by MaxVer, three units by neural networks and four units with similar accuracies in both approach classification. The overall accuracy of soil maps made by the conventional method was 82%. The soils of the first component in mapping units agreed in 48% with reference soils. |
id |
UFV_84dc5132aaf3fbe0dd3d89bb3707b6ec |
---|---|
oai_identifier_str |
oai:locus.ufv.br:123456789/5456 |
network_acronym_str |
UFV |
network_name_str |
LOCUS Repositório Institucional da UFV |
repository_id_str |
|
spelling |
Souza, Eliana dehttp://lattes.cnpq.br/9050394316046141Ker, João Carloshttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4763842Z5Schaefer, Carlos Ernesto Gonçalves Reynaudhttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4723204Y8Fernandes Filho, Elpídio Ináciohttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4703656Z4Vieira, Carlos Antonio Oliveirahttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4728250D0Chagas, César da Silvahttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4785135D2Simas, Felipe Nogueira Bellohttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4766936J52015-03-26T13:53:16Z2011-12-122015-03-26T13:53:16Z2009-02-16SOUZA, Eliana de. Supervised classification of soils using artificial neural networks in the Serra do Cipó - MG. 2009. 112 f. Dissertação (Mestrado em Fertilidade do solo e nutrição de plantas; Gênese, Morfologia e Classificação, Mineralogia, Química,) - Universidade Federal de Viçosa, Viçosa, 2009.http://locus.ufv.br/handle/123456789/5456The supervised classification of soils, especially in recent decades, is being carried out using mathematical and statistical models, amongst which the model of neural networks, stand out by greater accuracy of mapping comparing to classical models, such as Maximum Likelihood (MaxVer), helping the conventional method of mapping. Neural Networks model has been performed mostly for soil properties, with little application for soil classes. This work aimed to undertake the classification of soil using neural networks and MaxVer for an area located in Serra do Cipó, in the State of Minas Gerais. The map units was defined based on information from 55 soil profiles classified accord to the Brazilian System of Soils Classification to the fourth categorical level, whereas compound units were done with basis on the similarity of soil properties and the characteristics of physics environment. The discriminates variables used included: six scenes of Landsat satellite image sensor ETM+, four indexes derived from this image (Clay minerals, Ferrous minerals, Iron oxide and NDVI), Digital Elevation Model and derived attributes: altitude, slope, compound topographic index, aspect, solar radiation, curvature and elevation amplitude, in addiction to geological and soil maps. Several variables combinations were tested in both classifiers, selecting those that best contribute to classify the soil with high accuracy on the two supervised mapping approaches. The classification by neural networks was performed using the Stuttgart Neural Network Simulator and the backpropagation algorithm, the framework and classification parameters were selected by training and statistical tests. The results obtained with both classifiers, neural networks and MaxVer were compared using ground data as reference. The same set of reference points was used to validate the soil map obtained by the conventional method of mapping. Maps obtained by the two classifiers using the group of variables that provided the best performance to the classification showed a good accuracy index, with no statistical difference in overall accuracy of the maps. The map generated by MaxVer showed a kappa index of 0.58, while the map from neural network showed an index of 0.60. Although the accuracy of the two maps was statistically similar, the classifiers efficiency in individual discrimination of soil units differed significantly, with two units being best classified by MaxVer, three units by neural networks and four units with similar accuracies in both approach classification. The overall accuracy of soil maps made by the conventional method was 82%. The soils of the first component in mapping units agreed in 48% with reference soils.A classificação supervisionada de solos, especialmente nas últimas décadas, vem sendo realizada com o auxílio de modelos matemáticos e estatísticos, dentre os quais destaca-se o modelo de redes neurais, o qual tem apresentado exatidão superior quando comparado com métodos clássicos, como o de Máxima Verossimilhança (MaxVer), auxiliando no método convencional de mapeamento. No entanto, na maioria dos trabalhos foram avaliadas as propriedades dos solos, sendo o estudo das classes de solos ainda incipiente. Assim, este trabalho teve como objetivo realizar a classificação de solos por redes neurais e pelo MaxVer para uma área situada na Serra do Cipó, no estado de Minas Gerais. Para tanto, utilizaram-se informações analíticas de 55 perfis de solos, classificados até o quarto nível categórico do Sistema Brasileiro de Classificação de Solos. As unidades do mapa de solos foram compostas por semelhanças entre as propriedades físicas do solo e as características do ambiente. As variáveis discriminantes avaliadas na classificação foram seis cenas da imagem do satélite Landsat, sensor ETM+; quatro índices derivados dessa imagem (Clay minerals, Ferrous minerals,Iron oxide e NDVI); modelo digital de elevação e atributos derivados: altitude, declividade, índice topográfico combinado, face de exposição, radiação solar, curvatura e amplitude altimétrica, além dos mapas geológico e pedológico. A partir desse conjunto de variáveis, identificaram-se aquelas que melhor contribuíram na discriminação dos solos, em cada uma das duas abordagens empregadas. Na classificação pelas redes neurais foram empregados o simulador Stuttgart Neural Network Simulator e o algoritmo backpropagation, sendo a arquitetura e os parâmetros selecionados por meio de tentativas e testes de significância estatística. Os resultados obtidos por ambos os classificadores, redes neurais e MaxVer, foram comparados entre si, utilizando-se a validação dos mapas com pontos de referência terrestre. Os mesmos pontos de referência foram utilizados para validar o mapa de solos obtido pelo método convencional de mapeamento. Os mapas obtidos pelos dois classificadores, utilizando o conjunto de varáveis que proporcionou melhor desempenho do classificador, apresentaram índice de exatidão considerado bom, sem diferença estatística na exatidão global dos mapas. O mapa melhor classificado pelo MaxVer apresentou índice kappa de 0,58, enquanto que, pelas redes neurais, o maior índice foi de 0,60. Esses valores não diferiram estatisticamente, entretanto, os classificadores diferiram na discriminação das unidades de solo, sendo duas unidades melhor classificadas pelo MaxVer, três pelas redes neurais e quatro unidades com exatidão estatisticamente igual para os dois classificadores. A exatidão global do mapa obtido pelo método convencional de mapeamento foi de 82%, sendo esse índice calculado pelo somatório dos solos de referência concordantes com qualquer componente da unidade. Os solos no primeiro componente das unidades de mapeamento apresentaram 48% de concordância com solos de referência.Coordenação de Aperfeiçoamento de Pessoal de Nível Superiorapplication/pdfporUniversidade Federal de ViçosaMestrado em Solos e Nutrição de PlantasUFVBRFertilidade do solo e nutrição de plantas; Gênese, Morfologia e Classificação, Mineralogia, Química,Mapemaneto digital de solosPedologiaRedes neurais artificiaisDigital mapping of soilPedologyArtificial neural networksCNPQ::CIENCIAS AGRARIAS::AGRONOMIA::CIENCIA DO SOLOClassificação supervisionada de solos por redes neurais artificiais na Serra do Cipó - MGSupervised classification of soils using artificial neural networks in the Serra do Cipó - MGinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisinfo:eu-repo/semantics/openAccessreponame:LOCUS Repositório Institucional da UFVinstname:Universidade Federal de Viçosa (UFV)instacron:UFVORIGINALtexto completo.pdfapplication/pdf3045713https://locus.ufv.br//bitstream/123456789/5456/1/texto%20completo.pdf29d033208d3f70ac0ab24c206abd28f5MD51TEXTtexto completo.pdf.txttexto completo.pdf.txtExtracted texttext/plain185145https://locus.ufv.br//bitstream/123456789/5456/2/texto%20completo.pdf.txt76e3b9a52d63af3f76964e493c5f9caaMD52THUMBNAILtexto completo.pdf.jpgtexto completo.pdf.jpgIM Thumbnailimage/jpeg3560https://locus.ufv.br//bitstream/123456789/5456/3/texto%20completo.pdf.jpga0da52a8cba1c61f94945f5b2100ceefMD53123456789/54562016-04-11 23:02:37.48oai:locus.ufv.br:123456789/5456Repositório InstitucionalPUBhttps://www.locus.ufv.br/oai/requestfabiojreis@ufv.bropendoar:21452016-04-12T02:02:37LOCUS Repositório Institucional da UFV - Universidade Federal de Viçosa (UFV)false |
dc.title.por.fl_str_mv |
Classificação supervisionada de solos por redes neurais artificiais na Serra do Cipó - MG |
dc.title.alternative.eng.fl_str_mv |
Supervised classification of soils using artificial neural networks in the Serra do Cipó - MG |
title |
Classificação supervisionada de solos por redes neurais artificiais na Serra do Cipó - MG |
spellingShingle |
Classificação supervisionada de solos por redes neurais artificiais na Serra do Cipó - MG Souza, Eliana de Mapemaneto digital de solos Pedologia Redes neurais artificiais Digital mapping of soil Pedology Artificial neural networks CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::CIENCIA DO SOLO |
title_short |
Classificação supervisionada de solos por redes neurais artificiais na Serra do Cipó - MG |
title_full |
Classificação supervisionada de solos por redes neurais artificiais na Serra do Cipó - MG |
title_fullStr |
Classificação supervisionada de solos por redes neurais artificiais na Serra do Cipó - MG |
title_full_unstemmed |
Classificação supervisionada de solos por redes neurais artificiais na Serra do Cipó - MG |
title_sort |
Classificação supervisionada de solos por redes neurais artificiais na Serra do Cipó - MG |
author |
Souza, Eliana de |
author_facet |
Souza, Eliana de |
author_role |
author |
dc.contributor.authorLattes.por.fl_str_mv |
http://lattes.cnpq.br/9050394316046141 |
dc.contributor.author.fl_str_mv |
Souza, Eliana de |
dc.contributor.advisor-co1.fl_str_mv |
Ker, João Carlos |
dc.contributor.advisor-co1Lattes.fl_str_mv |
http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4763842Z5 |
dc.contributor.advisor-co2.fl_str_mv |
Schaefer, Carlos Ernesto Gonçalves Reynaud |
dc.contributor.advisor-co2Lattes.fl_str_mv |
http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4723204Y8 |
dc.contributor.advisor1.fl_str_mv |
Fernandes Filho, Elpídio Inácio |
dc.contributor.advisor1Lattes.fl_str_mv |
http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4703656Z4 |
dc.contributor.referee1.fl_str_mv |
Vieira, Carlos Antonio Oliveira |
dc.contributor.referee1Lattes.fl_str_mv |
http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4728250D0 |
dc.contributor.referee2.fl_str_mv |
Chagas, César da Silva |
dc.contributor.referee2Lattes.fl_str_mv |
http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4785135D2 |
dc.contributor.referee3.fl_str_mv |
Simas, Felipe Nogueira Bello |
dc.contributor.referee3Lattes.fl_str_mv |
http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4766936J5 |
contributor_str_mv |
Ker, João Carlos Schaefer, Carlos Ernesto Gonçalves Reynaud Fernandes Filho, Elpídio Inácio Vieira, Carlos Antonio Oliveira Chagas, César da Silva Simas, Felipe Nogueira Bello |
dc.subject.por.fl_str_mv |
Mapemaneto digital de solos Pedologia Redes neurais artificiais |
topic |
Mapemaneto digital de solos Pedologia Redes neurais artificiais Digital mapping of soil Pedology Artificial neural networks CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::CIENCIA DO SOLO |
dc.subject.eng.fl_str_mv |
Digital mapping of soil Pedology Artificial neural networks |
dc.subject.cnpq.fl_str_mv |
CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::CIENCIA DO SOLO |
description |
The supervised classification of soils, especially in recent decades, is being carried out using mathematical and statistical models, amongst which the model of neural networks, stand out by greater accuracy of mapping comparing to classical models, such as Maximum Likelihood (MaxVer), helping the conventional method of mapping. Neural Networks model has been performed mostly for soil properties, with little application for soil classes. This work aimed to undertake the classification of soil using neural networks and MaxVer for an area located in Serra do Cipó, in the State of Minas Gerais. The map units was defined based on information from 55 soil profiles classified accord to the Brazilian System of Soils Classification to the fourth categorical level, whereas compound units were done with basis on the similarity of soil properties and the characteristics of physics environment. The discriminates variables used included: six scenes of Landsat satellite image sensor ETM+, four indexes derived from this image (Clay minerals, Ferrous minerals, Iron oxide and NDVI), Digital Elevation Model and derived attributes: altitude, slope, compound topographic index, aspect, solar radiation, curvature and elevation amplitude, in addiction to geological and soil maps. Several variables combinations were tested in both classifiers, selecting those that best contribute to classify the soil with high accuracy on the two supervised mapping approaches. The classification by neural networks was performed using the Stuttgart Neural Network Simulator and the backpropagation algorithm, the framework and classification parameters were selected by training and statistical tests. The results obtained with both classifiers, neural networks and MaxVer were compared using ground data as reference. The same set of reference points was used to validate the soil map obtained by the conventional method of mapping. Maps obtained by the two classifiers using the group of variables that provided the best performance to the classification showed a good accuracy index, with no statistical difference in overall accuracy of the maps. The map generated by MaxVer showed a kappa index of 0.58, while the map from neural network showed an index of 0.60. Although the accuracy of the two maps was statistically similar, the classifiers efficiency in individual discrimination of soil units differed significantly, with two units being best classified by MaxVer, three units by neural networks and four units with similar accuracies in both approach classification. The overall accuracy of soil maps made by the conventional method was 82%. The soils of the first component in mapping units agreed in 48% with reference soils. |
publishDate |
2009 |
dc.date.issued.fl_str_mv |
2009-02-16 |
dc.date.available.fl_str_mv |
2011-12-12 2015-03-26T13:53:16Z |
dc.date.accessioned.fl_str_mv |
2015-03-26T13:53:16Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.citation.fl_str_mv |
SOUZA, Eliana de. Supervised classification of soils using artificial neural networks in the Serra do Cipó - MG. 2009. 112 f. Dissertação (Mestrado em Fertilidade do solo e nutrição de plantas; Gênese, Morfologia e Classificação, Mineralogia, Química,) - Universidade Federal de Viçosa, Viçosa, 2009. |
dc.identifier.uri.fl_str_mv |
http://locus.ufv.br/handle/123456789/5456 |
identifier_str_mv |
SOUZA, Eliana de. Supervised classification of soils using artificial neural networks in the Serra do Cipó - MG. 2009. 112 f. Dissertação (Mestrado em Fertilidade do solo e nutrição de plantas; Gênese, Morfologia e Classificação, Mineralogia, Química,) - Universidade Federal de Viçosa, Viçosa, 2009. |
url |
http://locus.ufv.br/handle/123456789/5456 |
dc.language.iso.fl_str_mv |
por |
language |
por |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Universidade Federal de Viçosa |
dc.publisher.program.fl_str_mv |
Mestrado em Solos e Nutrição de Plantas |
dc.publisher.initials.fl_str_mv |
UFV |
dc.publisher.country.fl_str_mv |
BR |
dc.publisher.department.fl_str_mv |
Fertilidade do solo e nutrição de plantas; Gênese, Morfologia e Classificação, Mineralogia, Química, |
publisher.none.fl_str_mv |
Universidade Federal de Viçosa |
dc.source.none.fl_str_mv |
reponame:LOCUS Repositório Institucional da UFV instname:Universidade Federal de Viçosa (UFV) instacron:UFV |
instname_str |
Universidade Federal de Viçosa (UFV) |
instacron_str |
UFV |
institution |
UFV |
reponame_str |
LOCUS Repositório Institucional da UFV |
collection |
LOCUS Repositório Institucional da UFV |
bitstream.url.fl_str_mv |
https://locus.ufv.br//bitstream/123456789/5456/1/texto%20completo.pdf https://locus.ufv.br//bitstream/123456789/5456/2/texto%20completo.pdf.txt https://locus.ufv.br//bitstream/123456789/5456/3/texto%20completo.pdf.jpg |
bitstream.checksum.fl_str_mv |
29d033208d3f70ac0ab24c206abd28f5 76e3b9a52d63af3f76964e493c5f9caa a0da52a8cba1c61f94945f5b2100ceef |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 |
repository.name.fl_str_mv |
LOCUS Repositório Institucional da UFV - Universidade Federal de Viçosa (UFV) |
repository.mail.fl_str_mv |
fabiojreis@ufv.br |
_version_ |
1794528593562304512 |