Depth-wise Squeeze and Excitation Block-based Efficient-Unet model for surface defect detection

Basit Öğe Kaydı
dc.authoridAslan, Muzaffer/0000-0002-2418-9472
dc.authoridHanbay, Davut/0000-0003-2271-7865
dc.authoridUZEN, Huseyin/0000-0002-0998-2130
dc.authorwosidAslan, Muzaffer/U-5355-2018
dc.authorwosidHanbay, Davut/AAG-8511-2019
dc.authorwosidUZEN, Huseyin/CZK-0841-2022
dc.contributor.authorUzen, Huseyin
dc.contributor.authorTurkoglu, Muammer
dc.contributor.authorAslan, Muzaffer
dc.contributor.authorHanbay, Davut
dc.date.accessioned2024-08-04T20:51:49Z
dc.date.available2024-08-04T20:51:49Z
dc.date.issued2023
dc.departmentİnönü Üniversitesien_US
dc.description.abstractDetection of surface defects in manufacturing systems is crucial for product quality. Detection of surface defects with high accuracy can prevent financial and time losses. Recently, efforts to develop high-performance automatic surface defect detection systems using computer vision and machine-learning methods have become prominent. In line with this purpose, this paper proposed a novel approach based on Depth-wise Squeeze and Excitation Block-based Efficient-Unet (DSEB-EUNet) for automatic surface defect detection. The proposed model consists of an encoder-decoder, the basic structure of the Unet architecture, and a Depth-wise Squeeze and Excitation Block added to the skip-connection of Unet. First, in the encoder part of the proposed model, low-level and high-level features were obtained by the EfficientNet network. Then, these features were transferred to the Depth-wise Squeeze and Excitation Block. The proposed DSEB based on the combination of Squeeze-Excitation and Depth-wise Separable Convolution enabled to reveal of critical information by weighting the features with a lightweight gating mechanism for surface defect detection. Besides, in the decoder part of the proposed model, the structure called Multi-level Feature Concatenated Block (MFCB) transferred the weighted features to the last layers without losing spatial detail. Finally, pixel-level defect detection was performed using the sigmoid function. The proposed model was tested using three general datasets for surface defect detection. In experimental works, the best F1-scores for MT, DAGM, and AITEX datasets using the proposed DSEB-EUNet architecture were 89.20%, 85.97%, and 90.39%, respectively. These results showed the proposed model outperforms higher performance compared to state-of-the-art approaches.en_US
dc.description.sponsorshipInonu University Scientific Research Projects Coordination [FDK-2021-2725]en_US
dc.description.sponsorshipThis work was supported by the Inonu University Scientific Research Projects Coordination [Grant Number FDK-2021-2725].en_US
dc.identifier.doi10.1007/s00371-022-02442-0
dc.identifier.endpage1764en_US
dc.identifier.issn0178-2789
dc.identifier.issn1432-2315
dc.identifier.issue5en_US
dc.identifier.scopus2-s2.0-85127403444en_US
dc.identifier.scopusqualityQ2en_US
dc.identifier.startpage1745en_US
dc.identifier.urihttps://doi.org/10.1007/s00371-022-02442-0
dc.identifier.urihttps://hdl.handle.net/11616/100581
dc.identifier.volume39en_US
dc.identifier.wosWOS:000776062200002en_US
dc.identifier.wosqualityQ2en_US
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.language.isoenen_US
dc.publisherSpringeren_US
dc.relation.ispartofVisual Computeren_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.subjectPixel-level surface defects' detectionen_US
dc.subjectConvolutional neural networken_US
dc.subjectDepth-wise Squeeze and Excitation Blocken_US
dc.subjectUneten_US
dc.titleDepth-wise Squeeze and Excitation Block-based Efficient-Unet model for surface defect detectionen_US
dc.typeArticleen_US

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu