000125361 001__ 125361
000125361 005__ 20251113150203.0
000125361 0247_ $$2doi$$a10.1109/TII.2023.3234030
000125361 0248_ $$2sideral$$a133106
000125361 037__ $$aART-2023-133106
000125361 041__ $$aspa
000125361 100__ $$aSampath, V.
000125361 245__ $$aAttention guided multi-task learning for surface defect identification
000125361 260__ $$c2023
000125361 5060_ $$aAccess copy available to the general public$$fUnrestricted
000125361 5203_ $$aSurface defect identification is an essential task in the industrial quality control process, in which visual checks are conducted on a manufactured product to ensure that it meets quality standards. Convolutional Neural Network (CNN) based surface defect identification method has proven to outperform traditional image processing techniques. However, the real-world surface defect datasets are limited in size due to the expensive data generation process and the rare occurrence of defects. To address this issue, this paper presents a method for exploiting auxiliary information beyond the primary labels to improve the generalization ability of surface defect identification tasks. Considering the correlation between pixel level segmentation masks, object level bounding boxes and global image level classification labels, we argue that jointly learning features of the related tasks can improve the performance of surface defect identification tasks. This paper proposes a framework named Defect-Aux-Net, based on multi-task learning with attention mechanisms that exploit the rich additional information from related tasks with the goal of simultaneously improving robustness and accuracy of the CNN based surface defect identification. We conducted a series of experiments with the proposed framework. The experimental results showed that the proposed method can significantly improve the performance of state-of-the-art models while achieving an overall accuracy of 97.1%, Dice score of 0.926 and mAP of 0.762 on defect classification, segmentation and detection tasks.
000125361 536__ $$9info:eu-repo/grantAgreement/EC/H2020/814225/EU/DIGItal MANufacturing Technologies for Zero-defect Industry 4.0 Production/DIGIMAN4.0$$9This project has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No H2020 814225-DIGIMAN4.0
000125361 540__ $$9info:eu-repo/semantics/openAccess$$aby$$uhttps://creativecommons.org/licenses/by/4.0/deed.es
000125361 590__ $$a11.7$$b2023
000125361 592__ $$a4.42$$b2023
000125361 591__ $$aAUTOMATION & CONTROL SYSTEMS$$b2 / 84 = 0.024$$c2023$$dQ1$$eT1
000125361 591__ $$aENGINEERING, INDUSTRIAL$$b2 / 69 = 0.029$$c2023$$dQ1$$eT1
000125361 591__ $$aCOMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS$$b3 / 170 = 0.018$$c2023$$dQ1$$eT1
000125361 593__ $$aComputer Science Applications$$c2023$$dQ1
000125361 593__ $$aInformation Systems$$c2023$$dQ1
000125361 593__ $$aElectrical and Electronic Engineering$$c2023$$dQ1
000125361 593__ $$aControl and Systems Engineering$$c2023$$dQ1
000125361 594__ $$a24.1$$b2023
000125361 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/publishedVersion
000125361 700__ $$aMaurtua, I.
000125361 700__ $$0(orcid)0000-0002-8609-1358$$aAguilar Martin, J. J.$$uUniversidad de Zaragoza
000125361 700__ $$aRivera, A.
000125361 700__ $$aMolina, J.
000125361 700__ $$aGutiérrez, A.
000125361 7102_ $$15002$$2515$$aUniversidad de Zaragoza$$bDpto. Ingeniería Diseño Fabri.$$cÁrea Ing. Procesos Fabricación
000125361 773__ $$g19, 9 (2023), 9713-9721$$pIEEE Trans. Ind. Inform.$$tIEEE Transactions on Industrial Informatics$$x1551-3203
000125361 8564_ $$s18338858$$uhttps://zaguan.unizar.es/record/125361/files/texto_completo.pdf$$yVersión publicada
000125361 8564_ $$s3541079$$uhttps://zaguan.unizar.es/record/125361/files/texto_completo.jpg?subformat=icon$$xicon$$yVersión publicada
000125361 909CO $$ooai:zaguan.unizar.es:125361$$particulos$$pdriver
000125361 951__ $$a2025-11-13-15:00:44
000125361 980__ $$aARTICLE