WMC-DFINE: An Improved DFINE Model for Aluminum Profile Surface Defect Detection
Sensors · 09.05.2026 · 26 (10) · 2994
DOI: 10.3390/s26102994 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 09.05.2026. Verfasst von Pengfei He und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Physics and Electronic Information, Yantai University, Yantai 264005, China. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: City 2023 School-Land Integration Development Project, Yantai Science and Technology-based Small and Medi-um-sized Enterprises Innovation Ability Enhancement Project Programme.
Abstract (Originalsprache)
The automated inspection of aluminum profile surface defects, which heavily relies on data acquired by machine vision sensors, is a critical task in industrial quality control. Addressing the current challenges of intense background texture interference and the difficulty in detecting defects with extreme aspect ratios on aluminum profiles, this research puts forward a complete end-to-end defect detection algorithm named WMC-DFINE (WIFA-MKSS-CSFF-DFINE) based on the DFINE framework. First, a Wavelet-Integrated Frequency Attention (WIFA) module is introduced, which utilizes a discrete wavelet transform to decouple features into the frequency domain, thereby dynamically suppressing high-frequency background noise and enhancing defect edge responses. Second, a Cross-Scale Feature Fusion (CSFF) module based on dual-channel pooling is designed to ensure the continuity of defect features, thereby resolving the semantic misalignment issue in traditional fusion. Third, a Multi-Kernel Strip Shuffle (MKSS) module is incorporated, utilizing decomposed convolution kernels to capture the geometric features of slender scratches. Finally, a knowledge distillation strategy is employed to transfer structured knowledge from a complex teacher model to a lightweight student model. Experiments on the Tianchi aluminum defect dataset demonstrate that WMC-DFINE achieves a mAP of 82.1%, which surpasses algorithms including YOLOv12, RT-DETR, and the baseline model DFINE. Furthermore, the distilled student model, WMC-DFINE-distill, improves the mAP by 3.2% compared to DFINE, reduces parameter count by 47%, and achieves an inference speed of 59.75 FPS on the experimental equipment. The proposed method effectively resolves the problem of balancing background suppression and defect detail feature preservation, offering a practical and efficient scheme for real-time industrial defect inspection.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Physics and Electronic Information, Yantai University, Yantai 264005, China.
- Shandong Provincial Laboratory of Data Open Innovation and Application for Advanced Smart Grid Technologies, Yantai University, Yantai 264005, China.
- Yantai Dongfang Wisdom Electric Co., Ltd., Yantai 264005, China.
ORCID-Profile: Pengfei He, Xia Liu
Förderung
- City 2023 School-Land Integration Development Project
- Yantai Science and Technology-based Small and Medi-um-sized Enterprises Innovation Ability Enhancement Project Programme
Zitieren
APA
He, P., Ding, Y., Yan, S., Wang, G., & Liu, X. (2026). WMC-DFINE: An Improved DFINE Model for Aluminum Profile Surface Defect Detection. Sensors, 26(10), 2994. https://doi.org/10.3390/s26102994
Vancouver
He P, Ding Y, Yan S, Wang G, Liu X. WMC-DFINE: An Improved DFINE Model for Aluminum Profile Surface Defect Detection. Sensors. 2026;26(10):2994. doi:10.3390/s26102994
BibTeX
@article{He2026,
author = {He, Pengfei and Ding, Yunming and Yan, Shuwen and Wang, Guoheng and Liu, Xia},
title = {WMC-DFINE: An Improved DFINE Model for Aluminum Profile Surface Defect Detection},
journal = {Sensors},
year = {2026},
volume = {26},
number = {10},
pages = {2994},
doi = {10.3390/s26102994}
}
EUVIMED