Development of a fast and cost-effective computed tomography system for industrial environments by incorporating priors into the imaging workflow
Autor(a) principal: | |
---|---|
Data de Publicação: | 2018 |
Tipo de documento: | Tese |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFPE |
Texto Completo: | https://repositorio.ufpe.br/handle/123456789/29992 |
Resumo: | Conventional X-ray radiography has been extensively used for inspection and quality assurance of industrial products. However, 2-D X-ray radiography cannot provide quantitative information within three dimensions about the scanned object. To obtain such depth information, X-ray Computed Tomography (CT) should be applied. Nevertheless, conventional CT systems (at which the X-ray source and detector rotates around the target object) are cost ineffective, inflexible, and suffer from long acquisition times. Therefore, the deployment of such technology is unfeasible for many industrial environments where high throughput is required as much as the best cost-benefit rate. The main goal of this research is to design a simple and cost-effective X-ray CT imaging system of high throughput for industrial environments. This system should comprises a single and static pair of X-ray source and detector for imaging objects passing on a conveyor belt. Such setup has been widely used with traditional radiographs for quality assurance in industrial environments; however, the large number of unknown projection views made such setup unfeasible for CT. Computer vision- and machine learning-based improvements are applied to incorporate prior knowledge about the scanned object into the CT imaging workflow as a way of compensating the lack of multiple X-ray sources or moving parts in both source and detector. More precisely, it is evaluated the use of priors related to the materials composition and also the outer object shape, as well as the use of Machine Learning techniques to apply priors automatically extracted from a training set of previous reconstructions. The trade-off between reconstruction quality and system’s throughput is exposed by linking the following measures: processing time, conveyor belt acceleration/deceleration, number of X-ray projections, reconstruction accuracy, and image resolution. It is also shown that one of the proposed methods can improve the system’s throughput in 21% while keeping the reconstruction accuracy over 90%. This research represents an advance in the state-of-the-art since it demonstrates that is possible to generate good quality reconstructions from projections acquired in an usual scanning setup where both X-ray source and detector are statically positioned. |
id |
UFPE_c1bb90cee247005673fc34cdf175d0a4 |
---|---|
oai_identifier_str |
oai:repositorio.ufpe.br:123456789/29992 |
network_acronym_str |
UFPE |
network_name_str |
Repositório Institucional da UFPE |
repository_id_str |
2221 |
spelling |
PEREIRA, Luis Filipe Alveshttp://lattes.cnpq.br/7320714889983490http://lattes.cnpq.br/8577312109146354CAVALCANTI, George Darmiton da CunhaREN, Tsang IngSIJBERS, Jan2019-04-02T20:48:16Z2019-04-02T20:48:16Z2018-03-20https://repositorio.ufpe.br/handle/123456789/29992Conventional X-ray radiography has been extensively used for inspection and quality assurance of industrial products. However, 2-D X-ray radiography cannot provide quantitative information within three dimensions about the scanned object. To obtain such depth information, X-ray Computed Tomography (CT) should be applied. Nevertheless, conventional CT systems (at which the X-ray source and detector rotates around the target object) are cost ineffective, inflexible, and suffer from long acquisition times. Therefore, the deployment of such technology is unfeasible for many industrial environments where high throughput is required as much as the best cost-benefit rate. The main goal of this research is to design a simple and cost-effective X-ray CT imaging system of high throughput for industrial environments. This system should comprises a single and static pair of X-ray source and detector for imaging objects passing on a conveyor belt. Such setup has been widely used with traditional radiographs for quality assurance in industrial environments; however, the large number of unknown projection views made such setup unfeasible for CT. Computer vision- and machine learning-based improvements are applied to incorporate prior knowledge about the scanned object into the CT imaging workflow as a way of compensating the lack of multiple X-ray sources or moving parts in both source and detector. More precisely, it is evaluated the use of priors related to the materials composition and also the outer object shape, as well as the use of Machine Learning techniques to apply priors automatically extracted from a training set of previous reconstructions. The trade-off between reconstruction quality and system’s throughput is exposed by linking the following measures: processing time, conveyor belt acceleration/deceleration, number of X-ray projections, reconstruction accuracy, and image resolution. It is also shown that one of the proposed methods can improve the system’s throughput in 21% while keeping the reconstruction accuracy over 90%. This research represents an advance in the state-of-the-art since it demonstrates that is possible to generate good quality reconstructions from projections acquired in an usual scanning setup where both X-ray source and detector are statically positioned.A radiografia tradicional, que utiliza raios-X, tem sido bastante utilizada para inspeção e controle de qualidade de produtos na indústria. No entanto, através de uma radiografia 2-D não é possível obter informações qualitativas em três dimensões sobre o objeto analisado. Para obter tal informação de profundidade, a Tomografia Computadorizada (TC) pode ser aplicada. Todavia, sistemas convencionais de TC (nos quais a fonte de raios-X e o detector giram em torno do objeto analisado) são de alto custo, inflexíveis, e necessitam de um longo tempo para aquisição de dados. Dessa forma, o uso dessa tecnologia é desaconselhável em ambientes industriais que demandam uma alta velocidade de processamento, além de baixo custo de implantação e manutenção. O principal objetivo dessa pesquisa é projetar um sistema de TC simples, de relativo baixo custo e de alta velocidade para ambientes industriais. Esse sistema deve ser composto por um único par de fonte e detector de raios-X posicionado estaticamente para escanear objetos que passam sobre uma esteira elétrica. Tal configuração tem sido extensamente utilizada para inspeção de qualidade utilizando radiografia 2-D em indústrias; no entanto, o baixo número de ângulos de visões disponíveis têm feito essa configuração inapropriada para tomografia. Técnicas implementadas usando visão computacional e aprendizagem de máquina são aplicadas para introduzir conhecimento a priori sobre o objeto em estudo no fluxo de dados da reconstrução de imagens em um sistema de TC, com isso espera-se compensar a falta de múltiplas fontes de raios-X ou movimentos entre a fonte e o detector de radiação. Mais precisamente, é avaliado o uso de conhecimento a priori sobre a composição do objeto e seu formato externo, bem como o uso de técnicas de Aprendizagem de Máquina para aplicar informações que foram extraídas automaticamente de um conjunto de treinamento formado por reconstruções anteriormente realizadas. O balanceamento entre a qualidade das reconstruções e a velocidade do sistema é apresentado nesse trabalho relacionando as seguintes medidas: tempo de processamento, aceleração/desaceleração da esteira, número de projeções de raios-X capturadas, acurácia da reconstrução e resolução da imagem reconstruída. Também é apresentado um método que é capaz de aumentar a velocidade do sistema em 21% enquanto a acurácia da reconstrução é mantida em ao menos 90%. A presente pesquisa contribuiu para o estado-da-arte da área ao demonstrar que é possível gerar reconstruções de boa qualidade a partir da aquisição de projeções em um sistema tomográfico não convencional no qual o emissor e o receptor de radiação estão posicionados estaticamente.engUniversidade Federal de PernambucoPrograma de Pos Graduacao em Ciencia da ComputacaoUFPEBrasilAttribution-NonCommercial-NoDerivs 3.0 Brazilhttp://creativecommons.org/licenses/by-nc-nd/3.0/br/info:eu-repo/semantics/openAccessCiência da computaçãoTomografia computadorizadaDevelopment of a fast and cost-effective computed tomography system for industrial environments by incorporating priors into the imaging workflowinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/doctoralThesisdoutoradoreponame:Repositório Institucional da UFPEinstname:Universidade Federal de Pernambuco (UFPE)instacron:UFPETHUMBNAILTESE Luis Filipe Alves Pereira.pdf.jpgTESE Luis Filipe Alves Pereira.pdf.jpgGenerated Thumbnailimage/jpeg1392https://repositorio.ufpe.br/bitstream/123456789/29992/6/TESE%20Luis%20Filipe%20Alves%20Pereira.pdf.jpgcf713ec39c9157bf673fe6bfe0d192a8MD56ORIGINALTESE Luis Filipe Alves Pereira.pdfTESE Luis Filipe Alves Pereira.pdfapplication/pdf2867511https://repositorio.ufpe.br/bitstream/123456789/29992/1/TESE%20Luis%20Filipe%20Alves%20Pereira.pdf231dd0b972125d6f83e6ee1aa6212cc1MD51LICENSElicense.txtlicense.txttext/plain; charset=utf-82311https://repositorio.ufpe.br/bitstream/123456789/29992/3/license.txt4b8a02c7f2818eaf00dcf2260dd5eb08MD53CC-LICENSElicense_rdflicense_rdfapplication/rdf+xml; charset=utf-8811https://repositorio.ufpe.br/bitstream/123456789/29992/4/license_rdfe39d27027a6cc9cb039ad269a5db8e34MD54TEXTTESE Luis Filipe Alves Pereira.pdf.txtTESE Luis Filipe Alves Pereira.pdf.txtExtracted texttext/plain197162https://repositorio.ufpe.br/bitstream/123456789/29992/5/TESE%20Luis%20Filipe%20Alves%20Pereira.pdf.txted5df74aab86c0e9dc3cab81e5a68b41MD55123456789/299922019-10-26 01:07:29.053oai:repositorio.ufpe.br:123456789/29992TGljZW7Dp2EgZGUgRGlzdHJpYnVpw6fDo28gTsOjbyBFeGNsdXNpdmEKClRvZG8gZGVwb3NpdGFudGUgZGUgbWF0ZXJpYWwgbm8gUmVwb3NpdMOzcmlvIEluc3RpdHVjaW9uYWwgKFJJKSBkZXZlIGNvbmNlZGVyLCDDoCBVbml2ZXJzaWRhZGUgRmVkZXJhbCBkZSBQZXJuYW1idWNvIChVRlBFKSwgdW1hIExpY2Vuw6dhIGRlIERpc3RyaWJ1acOnw6NvIE7Do28gRXhjbHVzaXZhIHBhcmEgbWFudGVyIGUgdG9ybmFyIGFjZXNzw612ZWlzIG9zIHNldXMgZG9jdW1lbnRvcywgZW0gZm9ybWF0byBkaWdpdGFsLCBuZXN0ZSByZXBvc2l0w7NyaW8uCgpDb20gYSBjb25jZXNzw6NvIGRlc3RhIGxpY2Vuw6dhIG7Do28gZXhjbHVzaXZhLCBvIGRlcG9zaXRhbnRlIG1hbnTDqW0gdG9kb3Mgb3MgZGlyZWl0b3MgZGUgYXV0b3IuCl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXwoKTGljZW7Dp2EgZGUgRGlzdHJpYnVpw6fDo28gTsOjbyBFeGNsdXNpdmEKCkFvIGNvbmNvcmRhciBjb20gZXN0YSBsaWNlbsOnYSBlIGFjZWl0w6EtbGEsIHZvY8OqIChhdXRvciBvdSBkZXRlbnRvciBkb3MgZGlyZWl0b3MgYXV0b3JhaXMpOgoKYSkgRGVjbGFyYSBxdWUgY29uaGVjZSBhIHBvbMOtdGljYSBkZSBjb3B5cmlnaHQgZGEgZWRpdG9yYSBkbyBzZXUgZG9jdW1lbnRvOwpiKSBEZWNsYXJhIHF1ZSBjb25oZWNlIGUgYWNlaXRhIGFzIERpcmV0cml6ZXMgcGFyYSBvIFJlcG9zaXTDs3JpbyBJbnN0aXR1Y2lvbmFsIGRhIFVGUEU7CmMpIENvbmNlZGUgw6AgVUZQRSBvIGRpcmVpdG8gbsOjbyBleGNsdXNpdm8gZGUgYXJxdWl2YXIsIHJlcHJvZHV6aXIsIGNvbnZlcnRlciAoY29tbyBkZWZpbmlkbyBhIHNlZ3VpciksIGNvbXVuaWNhciBlL291IGRpc3RyaWJ1aXIsIG5vIFJJLCBvIGRvY3VtZW50byBlbnRyZWd1ZSAoaW5jbHVpbmRvIG8gcmVzdW1vL2Fic3RyYWN0KSBlbSBmb3JtYXRvIGRpZ2l0YWwgb3UgcG9yIG91dHJvIG1laW87CmQpIERlY2xhcmEgcXVlIGF1dG9yaXphIGEgVUZQRSBhIGFycXVpdmFyIG1haXMgZGUgdW1hIGPDs3BpYSBkZXN0ZSBkb2N1bWVudG8gZSBjb252ZXJ0w6otbG8sIHNlbSBhbHRlcmFyIG8gc2V1IGNvbnRlw7pkbywgcGFyYSBxdWFscXVlciBmb3JtYXRvIGRlIGZpY2hlaXJvLCBtZWlvIG91IHN1cG9ydGUsIHBhcmEgZWZlaXRvcyBkZSBzZWd1cmFuw6dhLCBwcmVzZXJ2YcOnw6NvIChiYWNrdXApIGUgYWNlc3NvOwplKSBEZWNsYXJhIHF1ZSBvIGRvY3VtZW50byBzdWJtZXRpZG8gw6kgbyBzZXUgdHJhYmFsaG8gb3JpZ2luYWwgZSBxdWUgZGV0w6ltIG8gZGlyZWl0byBkZSBjb25jZWRlciBhIHRlcmNlaXJvcyBvcyBkaXJlaXRvcyBjb250aWRvcyBuZXN0YSBsaWNlbsOnYS4gRGVjbGFyYSB0YW1iw6ltIHF1ZSBhIGVudHJlZ2EgZG8gZG9jdW1lbnRvIG7Do28gaW5mcmluZ2Ugb3MgZGlyZWl0b3MgZGUgb3V0cmEgcGVzc29hIG91IGVudGlkYWRlOwpmKSBEZWNsYXJhIHF1ZSwgbm8gY2FzbyBkbyBkb2N1bWVudG8gc3VibWV0aWRvIGNvbnRlciBtYXRlcmlhbCBkbyBxdWFsIG7Do28gZGV0w6ltIG9zIGRpcmVpdG9zIGRlCmF1dG9yLCBvYnRldmUgYSBhdXRvcml6YcOnw6NvIGlycmVzdHJpdGEgZG8gcmVzcGVjdGl2byBkZXRlbnRvciBkZXNzZXMgZGlyZWl0b3MgcGFyYSBjZWRlciDDoApVRlBFIG9zIGRpcmVpdG9zIHJlcXVlcmlkb3MgcG9yIGVzdGEgTGljZW7Dp2EgZSBhdXRvcml6YXIgYSB1bml2ZXJzaWRhZGUgYSB1dGlsaXrDoS1sb3MgbGVnYWxtZW50ZS4gRGVjbGFyYSB0YW1iw6ltIHF1ZSBlc3NlIG1hdGVyaWFsIGN1am9zIGRpcmVpdG9zIHPDo28gZGUgdGVyY2Vpcm9zIGVzdMOhIGNsYXJhbWVudGUgaWRlbnRpZmljYWRvIGUgcmVjb25oZWNpZG8gbm8gdGV4dG8gb3UgY29udGXDumRvIGRvIGRvY3VtZW50byBlbnRyZWd1ZTsKZykgU2UgbyBkb2N1bWVudG8gZW50cmVndWUgw6kgYmFzZWFkbyBlbSB0cmFiYWxobyBmaW5hbmNpYWRvIG91IGFwb2lhZG8gcG9yIG91dHJhIGluc3RpdHVpw6fDo28gcXVlIG7Do28gYSBVRlBFLMKgZGVjbGFyYSBxdWUgY3VtcHJpdSBxdWFpc3F1ZXIgb2JyaWdhw6fDtWVzIGV4aWdpZGFzIHBlbG8gcmVzcGVjdGl2byBjb250cmF0byBvdSBhY29yZG8uCgpBIFVGUEUgaWRlbnRpZmljYXLDoSBjbGFyYW1lbnRlIG8ocykgbm9tZShzKSBkbyhzKSBhdXRvciAoZXMpIGRvcyBkaXJlaXRvcyBkbyBkb2N1bWVudG8gZW50cmVndWUgZSBuw6NvIGZhcsOhIHF1YWxxdWVyIGFsdGVyYcOnw6NvLCBwYXJhIGFsw6ltIGRvIHByZXZpc3RvIG5hIGFsw61uZWEgYykuCg==Repositório InstitucionalPUBhttps://repositorio.ufpe.br/oai/requestattena@ufpe.bropendoar:22212019-10-26T04:07:29Repositório Institucional da UFPE - Universidade Federal de Pernambuco (UFPE)false |
dc.title.pt_BR.fl_str_mv |
Development of a fast and cost-effective computed tomography system for industrial environments by incorporating priors into the imaging workflow |
title |
Development of a fast and cost-effective computed tomography system for industrial environments by incorporating priors into the imaging workflow |
spellingShingle |
Development of a fast and cost-effective computed tomography system for industrial environments by incorporating priors into the imaging workflow PEREIRA, Luis Filipe Alves Ciência da computação Tomografia computadorizada |
title_short |
Development of a fast and cost-effective computed tomography system for industrial environments by incorporating priors into the imaging workflow |
title_full |
Development of a fast and cost-effective computed tomography system for industrial environments by incorporating priors into the imaging workflow |
title_fullStr |
Development of a fast and cost-effective computed tomography system for industrial environments by incorporating priors into the imaging workflow |
title_full_unstemmed |
Development of a fast and cost-effective computed tomography system for industrial environments by incorporating priors into the imaging workflow |
title_sort |
Development of a fast and cost-effective computed tomography system for industrial environments by incorporating priors into the imaging workflow |
author |
PEREIRA, Luis Filipe Alves |
author_facet |
PEREIRA, Luis Filipe Alves |
author_role |
author |
dc.contributor.authorLattes.pt_BR.fl_str_mv |
http://lattes.cnpq.br/7320714889983490 |
dc.contributor.advisorLattes.pt_BR.fl_str_mv |
http://lattes.cnpq.br/8577312109146354 |
dc.contributor.author.fl_str_mv |
PEREIRA, Luis Filipe Alves |
dc.contributor.advisor1.fl_str_mv |
CAVALCANTI, George Darmiton da Cunha |
dc.contributor.advisor-co1.fl_str_mv |
REN, Tsang Ing SIJBERS, Jan |
contributor_str_mv |
CAVALCANTI, George Darmiton da Cunha REN, Tsang Ing SIJBERS, Jan |
dc.subject.por.fl_str_mv |
Ciência da computação Tomografia computadorizada |
topic |
Ciência da computação Tomografia computadorizada |
description |
Conventional X-ray radiography has been extensively used for inspection and quality assurance of industrial products. However, 2-D X-ray radiography cannot provide quantitative information within three dimensions about the scanned object. To obtain such depth information, X-ray Computed Tomography (CT) should be applied. Nevertheless, conventional CT systems (at which the X-ray source and detector rotates around the target object) are cost ineffective, inflexible, and suffer from long acquisition times. Therefore, the deployment of such technology is unfeasible for many industrial environments where high throughput is required as much as the best cost-benefit rate. The main goal of this research is to design a simple and cost-effective X-ray CT imaging system of high throughput for industrial environments. This system should comprises a single and static pair of X-ray source and detector for imaging objects passing on a conveyor belt. Such setup has been widely used with traditional radiographs for quality assurance in industrial environments; however, the large number of unknown projection views made such setup unfeasible for CT. Computer vision- and machine learning-based improvements are applied to incorporate prior knowledge about the scanned object into the CT imaging workflow as a way of compensating the lack of multiple X-ray sources or moving parts in both source and detector. More precisely, it is evaluated the use of priors related to the materials composition and also the outer object shape, as well as the use of Machine Learning techniques to apply priors automatically extracted from a training set of previous reconstructions. The trade-off between reconstruction quality and system’s throughput is exposed by linking the following measures: processing time, conveyor belt acceleration/deceleration, number of X-ray projections, reconstruction accuracy, and image resolution. It is also shown that one of the proposed methods can improve the system’s throughput in 21% while keeping the reconstruction accuracy over 90%. This research represents an advance in the state-of-the-art since it demonstrates that is possible to generate good quality reconstructions from projections acquired in an usual scanning setup where both X-ray source and detector are statically positioned. |
publishDate |
2018 |
dc.date.issued.fl_str_mv |
2018-03-20 |
dc.date.accessioned.fl_str_mv |
2019-04-02T20:48:16Z |
dc.date.available.fl_str_mv |
2019-04-02T20:48:16Z |
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.uri.fl_str_mv |
https://repositorio.ufpe.br/handle/123456789/29992 |
url |
https://repositorio.ufpe.br/handle/123456789/29992 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
Attribution-NonCommercial-NoDerivs 3.0 Brazil http://creativecommons.org/licenses/by-nc-nd/3.0/br/ info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Attribution-NonCommercial-NoDerivs 3.0 Brazil http://creativecommons.org/licenses/by-nc-nd/3.0/br/ |
eu_rights_str_mv |
openAccess |
dc.publisher.none.fl_str_mv |
Universidade Federal de Pernambuco |
dc.publisher.program.fl_str_mv |
Programa de Pos Graduacao em Ciencia da Computacao |
dc.publisher.initials.fl_str_mv |
UFPE |
dc.publisher.country.fl_str_mv |
Brasil |
publisher.none.fl_str_mv |
Universidade Federal de Pernambuco |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UFPE instname:Universidade Federal de Pernambuco (UFPE) instacron:UFPE |
instname_str |
Universidade Federal de Pernambuco (UFPE) |
instacron_str |
UFPE |
institution |
UFPE |
reponame_str |
Repositório Institucional da UFPE |
collection |
Repositório Institucional da UFPE |
bitstream.url.fl_str_mv |
https://repositorio.ufpe.br/bitstream/123456789/29992/6/TESE%20Luis%20Filipe%20Alves%20Pereira.pdf.jpg https://repositorio.ufpe.br/bitstream/123456789/29992/1/TESE%20Luis%20Filipe%20Alves%20Pereira.pdf https://repositorio.ufpe.br/bitstream/123456789/29992/3/license.txt https://repositorio.ufpe.br/bitstream/123456789/29992/4/license_rdf https://repositorio.ufpe.br/bitstream/123456789/29992/5/TESE%20Luis%20Filipe%20Alves%20Pereira.pdf.txt |
bitstream.checksum.fl_str_mv |
cf713ec39c9157bf673fe6bfe0d192a8 231dd0b972125d6f83e6ee1aa6212cc1 4b8a02c7f2818eaf00dcf2260dd5eb08 e39d27027a6cc9cb039ad269a5db8e34 ed5df74aab86c0e9dc3cab81e5a68b41 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional da UFPE - Universidade Federal de Pernambuco (UFPE) |
repository.mail.fl_str_mv |
attena@ufpe.br |
_version_ |
1802310612387102720 |