Close
Metadata

1. Identity statement
Reference TypeReport
Sitemtc-m21b.sid.inpe.br
Holder Codeisadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S
Identifier8JMKD3MGP3W34P/3PDRJN2
Repositorysid.inpe.br/mtc-m21b/2017/08.09.15.45   (restricted access)
Last Update2017:08.09.15.45.40 (UTC) marcelo.pazos@inpe.br
Metadata Repositorysid.inpe.br/mtc-m21b/2017/08.09.15.45.40
Metadata Last Update2018:06.04.03.24.45 (UTC) administrator
Report NumberINPE-1390-PE/182
Citation KeyKumarNier:1978:ClAcDi
TitleClassification accuracy of different options of the image-100 system
ProjectDSE/DIN
Year1978
Access Date2024, May 05
TypeRPQ
Number of Pages16
Number of Files1
Size5590 KiB
2. Context
Author1 Kumar, R.
2 Niero, M.
Group1 INPE-BR
2 INPE-BR
Affiliation1 Instituto Nacional de Pesquisas Espaciais (INPE)
2 Instituto Nacional de Pesquisas Espaciais (INPE)
e-Mail Addressmarcelo.pazos@inpe.br
InstitutionInstituto Nacional de Pesquisas Espaciais
CitySão José dos Campos
History (UTC)2017-08-09 15:47:39 :: marcelo.pazos@inpe.br :: -> 1978
2017-08-09 15:49:03 :: marcelo.pazos@inpe.br -> administrator :: 1978
2018-06-04 03:24:45 :: administrator -> marcelo.pazos@inpe.br :: 1978
3. Content and structure
Is the master or a copy?is the master
Content Stagecompleted
Transferable1
Keywordsland use
classification accuracy
remote sensing
AbstractThe purpose of this study was to compare the classification accuracy of land use classes of São Jose dos Campos, SP, Brazil, using the different options of signature acquisition for classification available in the Image-100, a system developed by General Electric Co., pixel-by-pixel maximum likelihood gaussian classifier (MLGC), and the sample classifier. In addition, the statistical separability of land use classes in the subsets of one to four spectral channels was investigated. With the help of ground observations and aerial photography, the multispectral scanner (MSS) data of LANDSAT were analysed using the Image-100. For the single -cell option of the Image-100, the errors of omission varied from 16.3% for the class "commercial" to 26.8% for the class "residential". The errors of commission varied from 5.6% for the class "commercial" to 33.2% for the class "Unoccupied". As expected, the multi-cell option increased the errors of omission and decreased the errors of commission. On the whole, the sample classifier gave slightly more accurate results than MLGC and much more accurate than any of the options of classification available in Image-100.
AreaETES
doc Directory Contentaccess
source Directory Contentthere are no files
agreement Directory Content
agreement.html 09/08/2017 12:45 1.7 KiB 
4. Conditions of access and use
Languagept
Target Filepublicacao.pdf
User Groupmarcelo.pazos@inpe.br
Visibilityshown
Copy HolderSID/SCD
Read Permissiondeny from all and allow from 150.163
Update Permissionnot transferred
5. Allied materials
Mirror Repositorysid.inpe.br/mtc-m21b/2013/09.26.14.25.22
DisseminationBNDEPOSITOLEGAL
Host Collectionsid.inpe.br/mtc-m21b/2013/09.26.14.25.20
6. Notes
Empty Fieldsarchivingpolicy archivist callnumber contenttype copyright creatorhistory date descriptionlevel doi edition electronicmailaddress format isbn issn label lineage mark nextedition nexthigherunit notes orcid parameterlist parentrepositories previousedition previouslowerunit progress readergroup recipient resumeid rightsholder schedulinginformation secondarydate secondarykey secondarymark secondarytype session shorttitle sponsor subject tertiarymark tertiarytype translator url versiontype
7. Description control
e-Mail (login)marcelo.pazos@inpe.br
update 


Close