<?xml version="1.0" encoding="ISO-8859-1"?>
<metadatalist>
	<metadata ReferenceType="Conference Proceedings">
		<site>marte2.sid.inpe.br 802</site>
		<holdercode>{isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S}</holdercode>
		<identifier>8JMKD3MGP6W34M/3PS4GEJ</identifier>
		<repository>sid.inpe.br/marte2/2017/10.23.19.33.52</repository>
		<lastupdate>2017:10.23.19.33.52 dpi.inpe.br/marte2/2013/05.17.15.03.06 banon</lastupdate>
		<metadatarepository>sid.inpe.br/marte2/2017/10.23.19.33.53</metadatarepository>
		<metadatalastupdate>2018:06.06.03.03.55 dpi.inpe.br/marte2/2013/05.17.15.03.06 administrator {D 2017}</metadatalastupdate>
		<isbn>978-85-17-00088-1</isbn>
		<label>59726</label>
		<citationkey>AlbuquerqueJardGianQuin:2017:InPrTo</citationkey>
		<title>Influence of pre-processing tools on the results of svm image classification for environmental monitoring</title>
		<format>Internet</format>
		<year>2017</year>
		<secondarytype>PRE CN</secondarytype>
		<numberoffiles>1</numberoffiles>
		<size>1294 KiB</size>
		<author>Albuquerque, Rafael Walter,</author>
		<author>Jardini, Mauricio George Miguel,</author>
		<author>Giannotti, Mariana Abrantes,</author>
		<author>Quintanilha, José Alberto,</author>
		<electronicmailaddress>r.w.albuquerque@gmail.com</electronicmailaddress>
		<editor>Gherardi, Douglas Francisco Marcolino,</editor>
		<editor>Aragão, Luiz Eduardo Oliveira e Cruz de,</editor>
		<e-mailaddress>daniela.seki@inpe.br</e-mailaddress>
		<conferencename>Simpósio Brasileiro de Sensoriamento Remoto, 18 (SBSR)</conferencename>
		<conferencelocation>Santos</conferencelocation>
		<date>28-31 maio 2017</date>
		<publisher>Instituto Nacional de Pesquisas Espaciais (INPE)</publisher>
		<publisheraddress>São José dos Campos</publisheraddress>
		<pages>1281-1288</pages>
		<booktitle>Anais</booktitle>
		<organization>Instituto Nacional de Pesquisas Espaciais (INPE)</organization>
		<transferableflag>1</transferableflag>
		<abstract>Hydroelectric power plants reservoirs perform over 80% of Brazilian energy sources, requiring sustainable policies to avoid expanding their land surface domain. Remote sensing (RS) satellite images provide a useful tool for monitoring land cover to help establish conservation policies on reservoir expansion. Unfortunately, these images, frequently offered by many companies, are already pre-processed to enhance visual attributes, which distorts radiometric content. This study aims to compare the SVM (Support Vector Machine) classification results on a satellite image that passed through two different treatments. The first treatment evolved a pure satellite image and the second a pre-processed image, both from the same satellite sensor and evolving the same study area. Labeling process performed by the SVM classifier on these two images showed a little difference between Kappa Indexes (variation of approximately 0.03), but Confusion Matrix showed significant differences in some land cover classes. To validate differences between the obtained classification results, a McNemar statistical test was applied and stated that accuracies were significantly different. Photointerpretation was also done to complement classification accuracy evaluation. Although only the automatic identification of houses, rocks and water showed less quantitative accurate results in the pure image, accuracy evaluation reinforced that the pre-processed image generally presented less accurate results. Therefore, these results should be taken into account by people who sell or buy satellite images with visual attributes enhanced.</abstract>
		<area>SRE</area>
		<type>Classificação e mineração de dados</type>
		<language>en</language>
		<targetfile>59726.pdf</targetfile>
		<usergroup>banon</usergroup>
		<visibility>shown</visibility>
		<mirrorrepository>urlib.net/www/2011/03.29.20.55</mirrorrepository>
		<nexthigherunit>8JMKD3MGP6W34M/3PMFNUS</nexthigherunit>
		<citingitemlist>sid.inpe.br/marte2/2017/09.25.14.55 3</citingitemlist>
		<hostcollection>dpi.inpe.br/marte2/2013/05.17.15.03.06</hostcollection>
		<username>banon</username>
		<lasthostcollection>dpi.inpe.br/marte2/2013/05.17.15.03.06</lasthostcollection>
		<url>http://marte2.sid.inpe.br/rep-/sid.inpe.br/marte2/2017/10.23.19.33.52</url>
	</metadata>
</metadatalist>