Mineração de dados, exatidão da classificação e modelagem do sombreamento do relevo no mapeamento do uso e cobertura da terra
Autor(a) principal: | |
---|---|
Data de Publicação: | 2013 |
Tipo de documento: | Tese |
Idioma: | por |
Título da fonte: | LOCUS Repositório Institucional da UFV |
Texto Completo: | http://locus.ufv.br/handle/123456789/1664 |
Resumo: | This work is segmented into three chapters. The first one deals with the quantification of the shadow areas in satellite images, established from existing data about the days and hours of satellite overpass, as well as the establishment of the best times for image acquisition in order to obtain the lowest percentage shadow. The end product of Remote Sensing featuring pixels with considerable areas of shade, especially in the high latitudes of the Northern and Southern Hemispheres. This effect highlights the landforms of relief, but in return affect the work of image classification, preventing it from being obtained class located below the shade. As yet it has not been possible to identify any data to present the area lost by shading in satellite images, it was decided to model the direct solar radiation that reaches the ground, date and time of passage of the Landsat TM and ETM +, whose images are used worldwide. For this, the same conditions were simulated relief at different latitudes, starting from 0° latitude (Ecuador) to 40° S latitude It was found that , in latitude 30° S and 40° S, where the area loss by shading goes from 27 % to 91 % , the images must be acquired, preferably between October and March. At latitudes 0° and 10 ° the loss can be considered negligible when a minimum threshold occurring in 10%. In the second chapter we used two Landsat TM images, acquired in one period of spring/winter, with extensive areas of shadows and over the period of the winter/fall, highlighting a bit of ground targets. These images were submitted to the process of data mining to obtain the best combination of bands and the maximum likelihood algorithm. Data mining was used to check whether there are combinations of bands that add better result in the classification, in addition to traditional 2,3,4 and 345, very widespread in Remote Sensing. In addition to the spectral bands of said sensor, three major components were used and the index NDVI total of 10 bands, combined together, resulting in 1,023 rankings for each image. The highest kappa obtained for the image of spring/winter was 0.90 with combination 235610 and for the winter/fall, up obteu kappa of 0.88 in combination 2358910. Were obtained conditional kappas lower classes only coffee and eucalyptus. When evaluating the effect of the set of bands was found that the main components and vegetation index NDVI increase in kappa provided. The NDVI appeared in all combinations with high kappas. The classifications that had at least one visible band, one of the infrared, one main component and NDVI, showed good results. In the last chapter, it was discussed the question of the accuracy of the maps produced, the indices used to assess the accuracy and limitations, as well as what has been proposed lately. For this, we used the results of the fifty best scores obtained with the two images that were evaluated, pixel by pixel. It was verified that the classes of coffee and eucalyptus were classified into seven to eight categories of diversity. While so much of the catchment area is classified in agreement, ie, these disagreements represent a smaller percentage. Nevertheless, the results showed that the statistical test employed z test, although widely used and recommended to assess statistical differences between binders or combinations of bands was not satisfactory for presenting different results, where, theoretically, there was no statistical difference in the z 5% level of probability. The results of the data showed no difference between responses showed accuracy of 41.4 % for the image of spring / winter and 60.5% for the image of the winter/fall. That is, based on the result obtained only area percentages of these possiam the same response in the classification process. |
id |
UFV_89db0826abbde97dcad93c6711894b55 |
---|---|
oai_identifier_str |
oai:locus.ufv.br:123456789/1664 |
network_acronym_str |
UFV |
network_name_str |
LOCUS Repositório Institucional da UFV |
repository_id_str |
2145 |
spelling |
França, Michelle Milanezhttp://lattes.cnpq.br/4699163871152127Fernandes Filho, Elpídio Ináciohttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4703656Z4Lani, João Luizhttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4783076P1Soares, Vicente Paulohttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4781715A9Ferreira, Williams Pinto Marqueshttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4799404E8Brandão, Pedro Christohttp://lattes.cnpq.br/68471199192933062015-03-26T12:52:57Z2014-11-032015-03-26T12:52:57Z2013-10-30FRANÇA, Michelle Milanez. Data mining, accuracy assessment and relief shading modeling on land use cover mapping. 2013. 104 f. Tese (Doutorado em Fertilidade do solo e nutrição de plantas; Gênese, Morfologia e Classificação, Mineralogia, Química,) - Universidade Federal de Viçosa, Viçosa, 2013.http://locus.ufv.br/handle/123456789/1664This work is segmented into three chapters. The first one deals with the quantification of the shadow areas in satellite images, established from existing data about the days and hours of satellite overpass, as well as the establishment of the best times for image acquisition in order to obtain the lowest percentage shadow. The end product of Remote Sensing featuring pixels with considerable areas of shade, especially in the high latitudes of the Northern and Southern Hemispheres. This effect highlights the landforms of relief, but in return affect the work of image classification, preventing it from being obtained class located below the shade. As yet it has not been possible to identify any data to present the area lost by shading in satellite images, it was decided to model the direct solar radiation that reaches the ground, date and time of passage of the Landsat TM and ETM +, whose images are used worldwide. For this, the same conditions were simulated relief at different latitudes, starting from 0° latitude (Ecuador) to 40° S latitude It was found that , in latitude 30° S and 40° S, where the area loss by shading goes from 27 % to 91 % , the images must be acquired, preferably between October and March. At latitudes 0° and 10 ° the loss can be considered negligible when a minimum threshold occurring in 10%. In the second chapter we used two Landsat TM images, acquired in one period of spring/winter, with extensive areas of shadows and over the period of the winter/fall, highlighting a bit of ground targets. These images were submitted to the process of data mining to obtain the best combination of bands and the maximum likelihood algorithm. Data mining was used to check whether there are combinations of bands that add better result in the classification, in addition to traditional 2,3,4 and 345, very widespread in Remote Sensing. In addition to the spectral bands of said sensor, three major components were used and the index NDVI total of 10 bands, combined together, resulting in 1,023 rankings for each image. The highest kappa obtained for the image of spring/winter was 0.90 with combination 235610 and for the winter/fall, up obteu kappa of 0.88 in combination 2358910. Were obtained conditional kappas lower classes only coffee and eucalyptus. When evaluating the effect of the set of bands was found that the main components and vegetation index NDVI increase in kappa provided. The NDVI appeared in all combinations with high kappas. The classifications that had at least one visible band, one of the infrared, one main component and NDVI, showed good results. In the last chapter, it was discussed the question of the accuracy of the maps produced, the indices used to assess the accuracy and limitations, as well as what has been proposed lately. For this, we used the results of the fifty best scores obtained with the two images that were evaluated, pixel by pixel. It was verified that the classes of coffee and eucalyptus were classified into seven to eight categories of diversity. While so much of the catchment area is classified in agreement, ie, these disagreements represent a smaller percentage. Nevertheless, the results showed that the statistical test employed z test, although widely used and recommended to assess statistical differences between binders or combinations of bands was not satisfactory for presenting different results, where, theoretically, there was no statistical difference in the z 5% level of probability. The results of the data showed no difference between responses showed accuracy of 41.4 % for the image of spring / winter and 60.5% for the image of the winter/fall. That is, based on the result obtained only area percentages of these possiam the same response in the classification process.Este trabalho está segmentado em três capítulos, inter-relacionados. O primeiro capítulo trata da quantificação das áreas de sombras em imagens de satélites, estabelecidos a partir de dados existentes acerca dos dias e horário de passagem do satélite, bem como do estabelecimento das melhores épocas para aquisição de imagens, de forma a obter o menor percentual de sombra. Os produtos do Sensoriamento Remoto acabam apresentando pixels com áreas consideráveis de sombra, principalmente nas altas latitudes dos Hemisférios Norte e Sul. Este efeito realça as geoformas do relevo, mas em contrapartida prejudica o trabalho de classificação de imagens, impedindo que seja obtida a classe situada abaixo da sombra. Como até o momento não foi possível identificar nenhum dado que apresente a área perdida por sombreamento em imagens de satélite, decidiu-se modelar a radiação solar direta que atinge o terreno, em data e horário de passagem do satélite Landsat TM e ETM+, cujas imagens são mundialmente utilizadas. Para isso, foram simuladas as mesmas condições de relevo, em latitudes distintas, partindo da latitude 0° (Equador) até a latitude 40° S. Verificou-se que, nas latitudes 30°S e 40°S, onde a perda de área por sombreamento vai de 27% a 91%, as imagens devem ser adquiridas, preferencialmente, entre outubro e março. Nas latitudes 0° e 10° a perda pode ser considerada desprezível, quando fixado um limiar mínimo de ocorrência em 10%. No segundo capítulo foram utilizadas duas imagens do satélite Landsat TM, uma adquirida em período da primavera/inverno, apresentando extensas áreas de sombras e outra do período do inverno/outono, realçando um pouco mais os alvos do terreno. Estas imagens foram submetidas ao processo de mineração de dados para obtenção da melhor combinação de bandas e o algoritmo Maxver. A mineração de dados foi utilizada para verificar se existem combinações de bandas que agregam melhor resultado final na classificação, além das tradicionais 2,3,4 e 345, muito difundidas no Sensoriamento Remoto. Além das bandas espectrais do referido sensor, foram utilizadas três componentes principais e o índice NDVI, totalizando 10 bandas que, combinadas entre si, resultaram em 1023 classificações para cada imagem. O maior índice kappa obtido para a imagem da primavera/inverno foi de 0,90 com a combinação 235610 e para o do inverno/outono, obteu-se kappa de 0,88 na combinação 2358910. Foram obtidos kappas condicionais baixos apenas para as classes café e eucalipto. Ao avaliar o efeito do conjunto de bandas verificou-se que as componentes principais e o índice de vegetação NDVI proporcionaram incremento no kappa. O NDVI apareceu em todas as combinações com kappas altos. As classificações que tiveram pelo menos uma banda do visível, uma do infra-vermelho, uma componente principal e o NDVI, apresentaram bons resultados. No último capítulo, foi abordado a questão da exatidão dos mapeamentos produzidos, os índices utilizados na avaliação da exatidão e suas limitações, bem como o que vem sendo proposto ultimamente. Para isso, foram utilizados os resultados das cinquenta melhores classificações obtidas com as duas imagens que foram avaliadas, pixel a pixel. Foi verificado que, as classes de café e eucalipto foram classificadas em sete a oito categorias de diversidade. Embora isso, grande parte da área da bacia é classificada em comum acordo, ou seja, essas discordâncias representam menor percentual. Mesmo assim, o resultado revelou que o teste estatístico empregado (teste z), embora muito utilizado e recomendado para avaliar diferenças estatísticas entre classificadores ou combinações de bandas, não foi satisfatório por apresentar resultados diferentes onde, teoricamente, não havia diferença estatística com z ao nível de 5% de probabilidade. Os resultados dos dados que não apresentaram divergência entre respostas mostraram acerto de 41,4% para a imagem do primavera/inverno e de 60,5% para a imagem da inverno/outono. Ou seja, com base no resultado obtido apenas estes percentuais de área possuam a mesma resposta no processo de classificação.Coordenação de Aperfeiçoamento de Pessoal de Nível Superiorapplication/pdfporUniversidade Federal de ViçosaDoutorado em Solos e Nutrição de PlantasUFVBRFertilidade do solo e nutrição de plantas; Gênese, Morfologia e Classificação, Mineralogia, Química,Solos - ClassificaçãoSensoriamento remotoProcessamento de imagensMineração de dados (Computação)Land- ClassificationRemote sensingImage processingData mining (Computing)CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::CIENCIA DO SOLOMineração de dados, exatidão da classificação e modelagem do sombreamento do relevo no mapeamento do uso e cobertura da terraData mining, accuracy assessment and relief shading modeling on land use cover mappinginfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/doctoralThesisinfo:eu-repo/semantics/openAccessreponame:LOCUS Repositório Institucional da UFVinstname:Universidade Federal de Viçosa (UFV)instacron:UFVORIGINALtexto completo.pdfapplication/pdf2776190https://locus.ufv.br//bitstream/123456789/1664/1/texto%20completo.pdf4bbee81c872b67a6d79b1cd184d99234MD51TEXTtexto completo.pdf.txttexto completo.pdf.txtExtracted texttext/plain191185https://locus.ufv.br//bitstream/123456789/1664/2/texto%20completo.pdf.txt4b2eb985618fc20d79a914c37507a424MD52THUMBNAILtexto completo.pdf.jpgtexto completo.pdf.jpgIM Thumbnailimage/jpeg3664https://locus.ufv.br//bitstream/123456789/1664/3/texto%20completo.pdf.jpg2086507e6e7c75a1297d67334fa1b64cMD53123456789/16642016-04-07 23:12:36.569oai:locus.ufv.br:123456789/1664Repositório InstitucionalPUBhttps://www.locus.ufv.br/oai/requestfabiojreis@ufv.bropendoar:21452016-04-08T02:12:36LOCUS Repositório Institucional da UFV - Universidade Federal de Viçosa (UFV)false |
dc.title.por.fl_str_mv |
Mineração de dados, exatidão da classificação e modelagem do sombreamento do relevo no mapeamento do uso e cobertura da terra |
dc.title.alternative.eng.fl_str_mv |
Data mining, accuracy assessment and relief shading modeling on land use cover mapping |
title |
Mineração de dados, exatidão da classificação e modelagem do sombreamento do relevo no mapeamento do uso e cobertura da terra |
spellingShingle |
Mineração de dados, exatidão da classificação e modelagem do sombreamento do relevo no mapeamento do uso e cobertura da terra França, Michelle Milanez Solos - Classificação Sensoriamento remoto Processamento de imagens Mineração de dados (Computação) Land- Classification Remote sensing Image processing Data mining (Computing) CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::CIENCIA DO SOLO |
title_short |
Mineração de dados, exatidão da classificação e modelagem do sombreamento do relevo no mapeamento do uso e cobertura da terra |
title_full |
Mineração de dados, exatidão da classificação e modelagem do sombreamento do relevo no mapeamento do uso e cobertura da terra |
title_fullStr |
Mineração de dados, exatidão da classificação e modelagem do sombreamento do relevo no mapeamento do uso e cobertura da terra |
title_full_unstemmed |
Mineração de dados, exatidão da classificação e modelagem do sombreamento do relevo no mapeamento do uso e cobertura da terra |
title_sort |
Mineração de dados, exatidão da classificação e modelagem do sombreamento do relevo no mapeamento do uso e cobertura da terra |
author |
França, Michelle Milanez |
author_facet |
França, Michelle Milanez |
author_role |
author |
dc.contributor.authorLattes.por.fl_str_mv |
http://lattes.cnpq.br/4699163871152127 |
dc.contributor.author.fl_str_mv |
França, Michelle Milanez |
dc.contributor.advisor-co1.fl_str_mv |
Fernandes Filho, Elpídio Inácio |
dc.contributor.advisor-co1Lattes.fl_str_mv |
http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4703656Z4 |
dc.contributor.advisor1.fl_str_mv |
Lani, João Luiz |
dc.contributor.advisor1Lattes.fl_str_mv |
http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4783076P1 |
dc.contributor.referee1.fl_str_mv |
Soares, Vicente Paulo |
dc.contributor.referee1Lattes.fl_str_mv |
http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4781715A9 |
dc.contributor.referee2.fl_str_mv |
Ferreira, Williams Pinto Marques |
dc.contributor.referee2Lattes.fl_str_mv |
http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4799404E8 |
dc.contributor.referee3.fl_str_mv |
Brandão, Pedro Christo |
dc.contributor.referee3Lattes.fl_str_mv |
http://lattes.cnpq.br/6847119919293306 |
contributor_str_mv |
Fernandes Filho, Elpídio Inácio Lani, João Luiz Soares, Vicente Paulo Ferreira, Williams Pinto Marques Brandão, Pedro Christo |
dc.subject.por.fl_str_mv |
Solos - Classificação Sensoriamento remoto Processamento de imagens Mineração de dados (Computação) |
topic |
Solos - Classificação Sensoriamento remoto Processamento de imagens Mineração de dados (Computação) Land- Classification Remote sensing Image processing Data mining (Computing) CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::CIENCIA DO SOLO |
dc.subject.eng.fl_str_mv |
Land- Classification Remote sensing Image processing Data mining (Computing) |
dc.subject.cnpq.fl_str_mv |
CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::CIENCIA DO SOLO |
description |
This work is segmented into three chapters. The first one deals with the quantification of the shadow areas in satellite images, established from existing data about the days and hours of satellite overpass, as well as the establishment of the best times for image acquisition in order to obtain the lowest percentage shadow. The end product of Remote Sensing featuring pixels with considerable areas of shade, especially in the high latitudes of the Northern and Southern Hemispheres. This effect highlights the landforms of relief, but in return affect the work of image classification, preventing it from being obtained class located below the shade. As yet it has not been possible to identify any data to present the area lost by shading in satellite images, it was decided to model the direct solar radiation that reaches the ground, date and time of passage of the Landsat TM and ETM +, whose images are used worldwide. For this, the same conditions were simulated relief at different latitudes, starting from 0° latitude (Ecuador) to 40° S latitude It was found that , in latitude 30° S and 40° S, where the area loss by shading goes from 27 % to 91 % , the images must be acquired, preferably between October and March. At latitudes 0° and 10 ° the loss can be considered negligible when a minimum threshold occurring in 10%. In the second chapter we used two Landsat TM images, acquired in one period of spring/winter, with extensive areas of shadows and over the period of the winter/fall, highlighting a bit of ground targets. These images were submitted to the process of data mining to obtain the best combination of bands and the maximum likelihood algorithm. Data mining was used to check whether there are combinations of bands that add better result in the classification, in addition to traditional 2,3,4 and 345, very widespread in Remote Sensing. In addition to the spectral bands of said sensor, three major components were used and the index NDVI total of 10 bands, combined together, resulting in 1,023 rankings for each image. The highest kappa obtained for the image of spring/winter was 0.90 with combination 235610 and for the winter/fall, up obteu kappa of 0.88 in combination 2358910. Were obtained conditional kappas lower classes only coffee and eucalyptus. When evaluating the effect of the set of bands was found that the main components and vegetation index NDVI increase in kappa provided. The NDVI appeared in all combinations with high kappas. The classifications that had at least one visible band, one of the infrared, one main component and NDVI, showed good results. In the last chapter, it was discussed the question of the accuracy of the maps produced, the indices used to assess the accuracy and limitations, as well as what has been proposed lately. For this, we used the results of the fifty best scores obtained with the two images that were evaluated, pixel by pixel. It was verified that the classes of coffee and eucalyptus were classified into seven to eight categories of diversity. While so much of the catchment area is classified in agreement, ie, these disagreements represent a smaller percentage. Nevertheless, the results showed that the statistical test employed z test, although widely used and recommended to assess statistical differences between binders or combinations of bands was not satisfactory for presenting different results, where, theoretically, there was no statistical difference in the z 5% level of probability. The results of the data showed no difference between responses showed accuracy of 41.4 % for the image of spring / winter and 60.5% for the image of the winter/fall. That is, based on the result obtained only area percentages of these possiam the same response in the classification process. |
publishDate |
2013 |
dc.date.issued.fl_str_mv |
2013-10-30 |
dc.date.available.fl_str_mv |
2014-11-03 2015-03-26T12:52:57Z |
dc.date.accessioned.fl_str_mv |
2015-03-26T12:52:57Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/doctoralThesis |
format |
doctoralThesis |
status_str |
publishedVersion |
dc.identifier.citation.fl_str_mv |
FRANÇA, Michelle Milanez. Data mining, accuracy assessment and relief shading modeling on land use cover mapping. 2013. 104 f. Tese (Doutorado em Fertilidade do solo e nutrição de plantas; Gênese, Morfologia e Classificação, Mineralogia, Química,) - Universidade Federal de Viçosa, Viçosa, 2013. |
dc.identifier.uri.fl_str_mv |
http://locus.ufv.br/handle/123456789/1664 |
identifier_str_mv |
FRANÇA, Michelle Milanez. Data mining, accuracy assessment and relief shading modeling on land use cover mapping. 2013. 104 f. Tese (Doutorado em Fertilidade do solo e nutrição de plantas; Gênese, Morfologia e Classificação, Mineralogia, Química,) - Universidade Federal de Viçosa, Viçosa, 2013. |
url |
http://locus.ufv.br/handle/123456789/1664 |
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 |
Doutorado 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/123456789/1664/1/texto%20completo.pdf https://locus.ufv.br//bitstream/123456789/1664/2/texto%20completo.pdf.txt https://locus.ufv.br//bitstream/123456789/1664/3/texto%20completo.pdf.jpg |
bitstream.checksum.fl_str_mv |
4bbee81c872b67a6d79b1cd184d99234 4b2eb985618fc20d79a914c37507a424 2086507e6e7c75a1297d67334fa1b64c |
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_ |
1801213011383287808 |