eBiltegia

    • Zer da eBiltegia? 
    •   eBiltegiari buruz
    •   Argitaratu irekian zure ikerketa
    • Sarbide Irekia MUn 
    •   Zer da Zientzia Irekia?
    •   Mondragon Unibertsitatearen dokumentu zientifikoetara eta irakaskuntza-materialetara Sarbide Irekia izateko politika instituzionala
    •   Mondragon Unibertsitatearen ikerketa-datuetara Sarbide Irekia izateko Politika instituzionala
    •   Babes digitalerako jarraibideak
    •   Zure argitalpenak jaso eta zabaldu egiten ditu Bibliotekak
    • Euskara
    • Español
    • English

Laguntzailea:

  • Kontaktua
  • Euskara 
    • Euskara
    • Español
    • English
  • eBiltegia buruz  
    • Zer da eBiltegia? 
    •   eBiltegiari buruz
    •   Argitaratu irekian zure ikerketa
    • Sarbide Irekia MUn 
    •   Zer da Zientzia Irekia?
    •   Mondragon Unibertsitatearen dokumentu zientifikoetara eta irakaskuntza-materialetara Sarbide Irekia izateko politika instituzionala
    •   Mondragon Unibertsitatearen ikerketa-datuetara Sarbide Irekia izateko Politika instituzionala
    •   Babes digitalerako jarraibideak
    •   Zure argitalpenak jaso eta zabaldu egiten ditu Bibliotekak
  • Hasi saioa
Ikusi itema 
  •   eBiltegia MONDRAGON UNIBERTSITATEA
  • Ikerketa-Kongresuak
  • Kongresuak-Ingeniaritza
  • Ikusi itema
  •   eBiltegia MONDRAGON UNIBERTSITATEA
  • Ikerketa-Kongresuak
  • Kongresuak-Ingeniaritza
  • Ikusi itema
JavaScript is disabled for your browser. Some features of this site may not work without it.
Ikusi/Ireki
Towards Robust Defect Detection in Casting Using Contrastive Learning.pdf (1.608Mb)
Erregistro osoa
Eragina

Web of Science   

Google Scholar
3
CITATIONS
3 total citations on Dimensions.
3 Total citations
3 Recent citations
1.84 Field Citation Ratio
n/a Relative Citation Ratio
Plum Print visual indicator of research metrics
plumX logo
  • Citations
    • Citation Indexes: 3
  • Captures
    • Readers: 4
see details
Partekatu
EmailLinkedinFacebookTwitter
Gorde erreferentzia
Mendely

Zotero

untranslated

Mets

Mods

Rdf

Marc

Exportar a BibTeX
Izenburua
Towards robust defect detection in casting using contrastive learning
Egilea
Intxausti Arbaiza, Eneko
Zugasti, Ekhi
Cernuda, Carlos
Egilea (beste erakunde batekoa)
Leibar, Ane Miren
Elizondo, Estibaliz
Ikerketa taldea
Análisis de datos y ciberseguridad
Beste instituzio
Fagor Ederlan, S. Coop.
Edertek S. Coop.
Bertsioa
Postprinta
Eskubideak
© 2023 Springer
Sarbidea
Sarbide bahitua
URI
https://hdl.handle.net/20.500.11984/6291
Argitaratzailearen bertsioa
https://doi.org/10.1007/978-3-031-49018-7_43
Non argitaratua
26th Iberoamerican Congress on Pattern Recognition (CIARP 2023). Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. Lecture Notes in Computer Science  Vol. 14469. Pp. 605-616.
Argitaratzailea
Springer
Gako-hitzak
Defect detection
contrastive learning
casting
optical quality control ... [+]
Defect detection
contrastive learning
casting
optical quality control
deep learning [-]
Laburpena
Defect detection plays a vital role in ensuring product quality and safety within industrial casting processes. In these dynamic environments, the occasional emergence of new defects in the production ... [+]
Defect detection plays a vital role in ensuring product quality and safety within industrial casting processes. In these dynamic environments, the occasional emergence of new defects in the production line poses a significant challenge for supervised methods. We present a defect detection framework to effectively detect novel defect patterns without prior exposure during training. Our method is based on contrastive learning applied to the Faster R-CNN model, enhanced with a contrastive head to obtain discriminative representations of different defects. By training on an diverse and comprehensive labeled dataset, our method achieves comparable performance to the supervised baseline model, showcasing commendable defect detection capabilities. To evaluate the robustness of our approach, we authentically replicate a real-world use case by deliberately excluding several defect types from the training data. Remarkably, in this new context, our proposed method significantly improves detection performance of the baseline model, particularly in situations with very limited training data, showcasing a remarkable 34.7% enhancement. Our research highlights the potential of the proposed method in real-world environments where the number of available images may be limited or inexistent. By providing valuable insights into defect detection in challenging scenarios, our framework could contribute to ensuring efficient and reliable product quality and safety in industrial manufacturing processes. [-]
Sponsorship
Gobierno Vasco
Finantzatzailea
Eusko Jaurlaritza = Gobierno Vasco
Programa
Elkartek 2022
Zenbakia
KK-2022/00049
Laguntzaren URIa
Sin información
Proiektua
Deeplearning REcomendation Manufacturing Imperfection Novelty Detection (DREMIND)
Bildumak
  • Kongresuak - Ingeniaritza [423]

Zerrendatu honako honen arabera

eBiltegia osoaKomunitateak & bildumakArgitalpen dataren araberaEgileakIzenburuakMateriakIkerketa taldeakNon argitaratuaBilduma hauArgitalpen dataren araberaEgileakIzenburuakMateriakIkerketa taldeakNon argitaratua

Nire kontua

SartuErregistratu

Estatistikak

Ikusi erabilearen inguruko estatistikak

Nork bildua:

OpenAIREBASERecolecta

Nork balioztatua:

OpenAIRERebiun
MONDRAGON UNIBERTSITATEA | Biblioteka
Kontaktua | Iradokizunak
DSpace
 

 

Nork bildua:

OpenAIREBASERecolecta

Nork balioztatua:

OpenAIRERebiun
MONDRAGON UNIBERTSITATEA | Biblioteka
Kontaktua | Iradokizunak
DSpace