Close

1. Identity statement
Reference TypeThesis or Dissertation (Thesis)
Sitemtc-m21c.sid.inpe.br
Holder Codeisadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S
Identifier8JMKD3MGP3W34R/3T2SPDL
Repositorysid.inpe.br/mtc-m21c/2019/03.29.13.33
Last Update2019:06.05.16.06.30 (UTC) simone
Metadata Repositorysid.inpe.br/mtc-m21c/2019/03.29.13.33.58
Metadata Last Update2019:07.03.10.15.03 (UTC) administrator
Secondary KeyINPE-18171-TDI/2859
Citation KeyCosta:2019:SeImSe
TitleSegmentação de imagens de sensoriamento remoto baseada em séries temporais e DTW
Alternate TitleSegmentation of remote sensing images based on time series and DTW
CourseCAP-COMP-SESPG-INPE-MCTIC-GOV-BR
Year2019
Date2019-02-28
Access Date2024, May 16
Thesis TypeTese (Doutorado em Computação Aplicada)
Secondary TypeTDI
Number of Pages125
Number of Files1
Size33241 KiB
2. Context
AuthorCosta, Wanderson Santos
CommitteeSantos, Rafael Duarte Coelho dos (presidente)
Fonseca, Leila Maria Garcia (orientadora)
Körting, Thales Sehn (orientador)
Sant'Anna, Sidnei João Siqueira
Happ, Patrick Nigri
Centeno, Jorge Antonio Silva
e-Mail Addresswscosta.inpe@gmail.com
UniversityInstituto Nacional de Pesquisas Espaciais (INPE)
CitySão José dos Campos
History (UTC)2019-03-29 13:33:58 :: wscosta.inpe@gmail.com -> pubtc@inpe.br ::
2019-03-29 18:50:44 :: pubtc@inpe.br -> wscosta.inpe@gmail.com ::
2019-04-25 00:54:27 :: wscosta.inpe@gmail.com -> pubtc@inpe.br ::
2019-06-06 16:42:03 :: pubtc@inpe.br -> administrator ::
2019-06-28 16:02:10 :: administrator -> simone ::
2019-06-28 16:04:08 :: simone :: -> 2019
2019-06-28 16:04:10 :: simone -> administrator :: 2019
2019-07-03 10:15:03 :: administrator -> :: 2019
3. Content and structure
Is the master or a copy?is the master
Content Stagecompleted
Transferable1
KeywordsSegmentação multitemporal
processamento de imagens
sensoriamento remoto
dynamic time warping
multitemporal segmentation
image processing
remote sensing
dynamic time warping
AbstractA disponibilidade de uma grande quantidade de dados de sensores remotos com diferentes resoluções temporais e espaciais tem tornado cada vez mais acessível e detalhada a observação da Terra. Dentro deste contexto, o uso de segmentadores eficientes em aplicações de sensoriamento remoto apresenta um papel importante neste cenário, ao buscar regiões homogêneas no domínio espaço-tempo e, consequentemente, reduzir o conjunto de dados. Além disso, a segmentação multitemporal pode trazer uma nova maneira de interpretação dos dados, ao produzir regiões contíguas no tempo. Portanto, este trabalho descreve um algoritmo de segmentação multitemporal baseado em séries temporais obtidas a partir de imagens ópticas de sensoriamento remoto. A distância Dynamic Time Warping foi utilizada como critério de homogeneidade na segmentação e quatro estudos de caso foram realizados para avaliar o método proposto. Nesta avaliação são usadas séries temporais de índices de vegetação NDVI e EVI geradas a partir de imagens MODIS, Landsat-8 e Landsat-7. Outros critérios de homogeneidade foram avaliados. As avaliações qualitativa e quantitativa demonstraram o potencial do método de segmentação proposto. ABSTRACT: The availability of a large amount of remote sensing data with different temporal and spatial resolutions has increasingly made Earth observation more accessible and detailed. In this context, the use of efficient remote sensing image segmenters in remote sensing applications plays an important role in this scenario when searching for homogeneous regions in space-time domain and, consequently, reducing the dataset. In addition, multitemporal segmentation can bring a new way of interpreting data, producing contiguous regions in time. Therefore, this thesis has the objective the development of a multitemporal segmentation algorithm based on time series from remote sensing optical images. The Dynamic Time Warping distance was used as the homogeneity criterion and four case studies were performed to evaluate the proposed method. In this evaluation, time series of vegetation indices NDVI and EVI were used, generated from MODIS, Landsat-8 and Landsat- 7 images. NDVI and EVI vegetation indices from these sensors were used to create the time series. Other homogeneity criteria were evaluated. The qualitative and quantitative evaluations demonstrated the potential of the proposed segmentation method.
AreaCOMP
Arrangementurlib.net > BDMCI > Fonds > Produção pgr ATUAIS > CAP > Segmentação de imagens...
doc Directory Contentaccess
source Directory Content
originais/@4primeirasPaginas.pdf 04/06/2019 08:43 176.9 KiB 
originais/Avaliação final pag 01 e 02 de Wanderson Santos Costa - CAP.pdf 05/06/2019 12:40 497.8 KiB 
originais/teseWandersonCorrecoesBancaBiblioteca.pdf 25/04/2019 15:51 31.9 MiB
agreement Directory Content
agreement.html 29/03/2019 10:33 1.7 KiB 
autorizacao.pdf 05/06/2019 13:06 1.0 MiB
4. Conditions of access and use
data URLhttp://urlib.net/ibi/8JMKD3MGP3W34R/3T2SPDL
zipped data URLhttp://urlib.net/zip/8JMKD3MGP3W34R/3T2SPDL
Languagept
Target Filepublicacao.pdf
User Groupgabinete@inpe.br
pubtc@inpe.br
wscosta.inpe@gmail.com
Visibilityshown
Copyright Licenseurlib.net/www/2012/11.12.15.10
Read Permissionallow from all
Update Permissionnot transferred
5. Allied materials
Mirror Repositoryurlib.net/www/2017/11.22.19.04.03
Next Higher Units8JMKD3MGPCW/3F2PHGS
Citing Item Listsid.inpe.br/bibdigital/2013/10.12.22.16 3
Host Collectionurlib.net/www/2017/11.22.19.04
6. Notes
Empty Fieldsacademicdepartment affiliation archivingpolicy archivist callnumber contenttype copyholder creatorhistory descriptionlevel dissemination doi electronicmailaddress format group isbn issn label lineage mark nextedition notes number orcid parameterlist parentrepositories previousedition previouslowerunit progress readergroup resumeid rightsholder schedulinginformation secondarydate secondarymark session shorttitle sponsor subject tertiarymark tertiarytype url versiontype


Close