<?xml version="1.0" encoding="ISO-8859-1"?>
<metadatalist>
	<metadata ReferenceType="Conference Proceedings">
		<site>sibgrapi.sid.inpe.br 802</site>
		<holdercode>{ibi 8JMKD3MGPEW34M/46T9EHH}</holdercode>
		<identifier>8JMKD3MGPAW/3PML3RP</identifier>
		<repository>sid.inpe.br/sibgrapi/2017/09.26.14.23</repository>
		<lastupdate>2017:09.26.14.23.27 sid.inpe.br/banon/2001/03.30.15.38 rafael89.rocha@gmail.com</lastupdate>
		<metadatarepository>sid.inpe.br/sibgrapi/2017/09.26.14.23.27</metadatarepository>
		<metadatalastupdate>2022:05.18.22.18.26 sid.inpe.br/banon/2001/03.30.15.38 administrator {D 2017}</metadatalastupdate>
		<citationkey>RochaSGSSRBDCS:2017:AvTéDe</citationkey>
		<title>Avaliação de técnicas de Deep Learning aplicadas à  identificação de peças defeituosas em vagões de trem</title>
		<format>On-line</format>
		<year>2017</year>
		<numberoffiles>1</numberoffiles>
		<size>2491 KiB</size>
		<author>Rocha, Rafael L.,</author>
		<author>Siravenha, Ana Carolina Q.,</author>
		<author>Gomes, Ana C. S.,</author>
		<author>Serejo, Gerson L.,</author>
		<author>Silva, Alexandre F. B.,</author>
		<author>Rodrigues, Luciano M.,</author>
		<author>Braga, Júlio,</author>
		<author>Dias, Giovanni,</author>
		<author>Carvalho, Schubert R.,</author>
		<author>Souza, Cleidson R. B. de,</author>
		<affiliation>Instituto Tecnológico Vale (ITV), Belém, Pará, Brasil</affiliation>
		<affiliation>Instituto Tecnológico Vale (ITV), Belém, Pará, Brasil</affiliation>
		<affiliation>Instituto SENAI de Inovação em Tecnologias Minerais ISI/SENAI, Belém, Pará, Brasil</affiliation>
		<affiliation>Instituto Tecnológico Vale (ITV), Belém, Pará, Brasil</affiliation>
		<affiliation>Instituto SENAI de Inovação em Tecnologias Minerais ISI/SENAI, Belém, Pará, Brasil</affiliation>
		<affiliation>Instituto Tecnológico Vale (ITV), Belém, Pará, Brasil</affiliation>
		<affiliation>Vale S.A. São Luís. MA, Brasil</affiliation>
		<affiliation>Vale S.A. São Luís. MA, Brasil</affiliation>
		<affiliation>Instituto Tecnológico Vale (ITV), Belém, Pará, Brasil</affiliation>
		<affiliation>Instituto Tecnológico Vale (ITV), Belém, Pará, Brasil e Universidade Federal do Pará, Belém, Pará, Brasil</affiliation>
		<editor>Torchelsen, Rafael Piccin,</editor>
		<editor>Nascimento, Erickson Rangel do,</editor>
		<editor>Panozzo, Daniele,</editor>
		<editor>Liu, Zicheng,</editor>
		<editor>Farias, Mylène,</editor>
		<editor>Viera, Thales,</editor>
		<editor>Sacht, Leonardo,</editor>
		<editor>Ferreira, Nivan,</editor>
		<editor>Comba, João Luiz Dihl,</editor>
		<editor>Hirata, Nina,</editor>
		<editor>Schiavon Porto, Marcelo,</editor>
		<editor>Vital, Creto,</editor>
		<editor>Pagot, Christian Azambuja,</editor>
		<editor>Petronetto, Fabiano,</editor>
		<editor>Clua, Esteban,</editor>
		<editor>Cardeal, Flávio,</editor>
		<e-mailaddress>rafael89.rocha@gmail.com</e-mailaddress>
		<conferencename>Conference on Graphics, Patterns and Images, 30 (SIBGRAPI)</conferencename>
		<conferencelocation>Niterói, RJ, Brazil</conferencelocation>
		<date>17-20 Oct. 2017</date>
		<publisher>Sociedade Brasileira de Computação</publisher>
		<publisheraddress>Porto Alegre</publisheraddress>
		<booktitle>Proceedings</booktitle>
		<tertiarytype>Industry Application Paper</tertiarytype>
		<transferableflag>1</transferableflag>
		<keywords>Deep learning, Convolutional neural network, Image classification, Inspection, Wagon components.</keywords>
		<abstract>Inspecting objects is an important task in many areas and is often used in industry to ensure product quality, allowing problem correction and disposal of damaged products. Inspection is also widely used in railway maintenance, where every day, hundreds of wagons are inspected visually in a process dependent on personal interpretation. This article describes an inspection approach of wagon components using deep learning techniques that comprises the stages of the component detection and the identification of its condition. In this work, the analyzed component is the shear pad which is responsible for supporting the truck. Object detection is done by a cascade detector and the classification task among three possible states (undamaged, absent and damaged) is done by convolutional neural networks. Our results are very encouraging, especially when observing the performance of the AlexNet network.</abstract>
		<language>pt</language>
		<targetfile>workshopsibgrapi2017.pdf</targetfile>
		<usergroup>rafael89.rocha@gmail.com</usergroup>
		<visibility>shown</visibility>
		<documentstage>not transferred</documentstage>
		<mirrorrepository>sid.inpe.br/banon/2001/03.30.15.38.24</mirrorrepository>
		<nexthigherunit>8JMKD3MGPAW/3PKCC58</nexthigherunit>
		<citingitemlist>sid.inpe.br/sibgrapi/2017/09.12.13.04 6</citingitemlist>
		<hostcollection>sid.inpe.br/banon/2001/03.30.15.38</hostcollection>
		<agreement>agreement.html .htaccess .htaccess2</agreement>
		<lasthostcollection>sid.inpe.br/banon/2001/03.30.15.38</lasthostcollection>
		<url>http://sibgrapi.sid.inpe.br/rep-/sid.inpe.br/sibgrapi/2017/09.26.14.23</url>
	</metadata>
</metadatalist>