Simple record

dc.rights.licenseAttribution 4.0 International*
dc.contributor.authorBalzategui, Julen
dc.contributor.authorEciolaza, Luka
dc.contributor.authorMaestro-Watson, Daniel
dc.date.accessioned2021-11-03T11:03:21Z
dc.date.available2021-11-03T11:03:21Z
dc.date.issued2021
dc.identifier.issn1424-8220en
dc.identifier.otherhttps://katalogoa.mondragon.edu/janium-bin/janium_login_opac.pl?find&ficha_no=164634en
dc.identifier.urihttps://hdl.handle.net/20.500.11984/5408
dc.description.abstractQuality inspection applications in industry are required to move towards a zero-defect manufacturing scenario, with non-destructive inspection and traceability of 100% of produced parts. Developing robust fault detection and classification models from the start-up of the lines is challenging due to the difficulty in getting enough representative samples of the faulty patterns and the need to manually label them. This work presents a methodology to develop a robust inspection system, targeting these peculiarities, in the context of solar cell manufacturing. The methodology is divided into two phases: In the first phase, an anomaly detection model based on a Generative Adversarial Network (GAN) is employed. This model enables the detection and localization of anomalous patterns within the solar cells from the beginning, using only non-defective samples for training and without any manual labeling involved. In a second stage, as defective samples arise, the detected anomalies will be used as automatically generated annotations for the supervised training of a Fully Convolutional Network that is capable of detecting multiple types of faults. The experimental results using 1873 Electroluminescence (EL) images of monocrystalline cells show that (a) the anomaly detection scheme can be used to start detecting features with very little available data, (b) the anomaly detection may serve as automatic labeling in order to train a supervised model, and (c) segmentation and classification results of supervised models trained with automatic labels are comparable to the ones obtained from the models trained with manual labels.en
dc.description.sponsorshipGobierno Vascoes
dc.language.isoengen
dc.publisherMDPIen
dc.rights© 2021 by the authors. Licensee MDPIen
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/*
dc.subjectanomaly detectionen
dc.subjectelectroluminescenceen
dc.subjectsolar cellsen
dc.subjectneural networksen
dc.titleAnomaly Detection and Automatic Labeling for Solar Cell Quality Inspection Based on Generative Adversarial Networken
dcterms.accessRightshttp://purl.org/coar/access_right/c_abf2en
dcterms.sourceSensorsen
local.contributor.groupRobótica y automatizaciónes
local.description.peerreviewedtrueen
local.identifier.doihttps://doi.org/10.3390/s21134361en
local.relation.projectIDinfo:eu-repo/grantAgreement/GV/Elkartek 2020/KK-2020-00077/CAPV/Desarrollo de tecnologías fotovoltaicas avanzadas/ENSOL2en
local.rights.publicationfeeAPCen
local.rights.publicationfeeamount1200 CHF 1137 EURen
local.source.detailsVol. 21. N. 13. N. articulo. 4361, 2021en
oaire.format.mimetypeapplication/pdf
oaire.file$DSPACE\assetstore
oaire.resourceTypehttp://purl.org/coar/resource_type/c_6501en
oaire.versionhttp://purl.org/coar/version/c_970fb48d4fbd8a85en


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Simple record

Attribution 4.0 International
Except where otherwise noted, this item's license is described as Attribution 4.0 International