Show simple item record

Optimized Adaptive Multi-Scale Architecture for Surface Defect Recognition

dc.contributor.authorXueli Chang
dc.contributor.authorYue Wang
dc.contributor.authorHeping Zhang
dc.contributor.authorBogdan Adamyk
dc.contributor.authorLingyu Yan
dc.contributor.otherSchool of Computer Science, Hubei University of Technology, Wuhan 430068, China
dc.contributor.otherSchool of Computer Science, Hubei University of Technology, Wuhan 430068, China
dc.contributor.otherSchool of Computer Science, Hubei University of Technology, Wuhan 430068, China
dc.contributor.otherAston Business School, Aston University, Birmingham B4 7ET, UK
dc.contributor.otherSchool of Computer Science, Hubei University of Technology, Wuhan 430068, China
dc.date.accessioned2025-08-27T14:04:00Z
dc.date.accessioned2025-10-08T08:44:56Z
dc.date.available2025-10-08T08:44:56Z
dc.date.issued01-08-2025
dc.identifier.urihttp://digilib.fisipol.ugm.ac.id/repo/handle/15717717/36973
dc.description.abstractDetection of defects on steel surface is crucial for industrial quality control. To address the issues of structural complexity, high parameter volume, and poor real-time performance in current detection models, this study proposes a lightweight model based on an improved YOLOv11. The model first reconstructs the backbone network by introducing a Reversible Connected Multi-Column Network (RevCol) to effectively preserve multi-level feature information. Second, the lightweight FasterNet is embedded into the C3k2 module, utilizing Partial Convolution (PConv) to reduce computational overhead. Additionally, a Group Convolution-driven EfficientDetect head is designed to maintain high-performance feature extraction while minimizing consumption of computational resources. Finally, a novel WISEPIoU loss function is developed by integrating WISE-IoU and POWERFUL-IoU to accelerate the model convergence and optimize the accuracy of bounding box regression. The experiments on the NEU-DET dataset demonstrate that the improved model achieves a parameter reduction of 39.1% from the baseline and computational complexity of 49.2% reduction in comparison with the baseline, with an mAP@0.5 of 0.758 and real-time performance of 91 FPS. On the DeepPCB dataset, the model exhibits reduction of parameters and computations by 39.1% and 49.2%, respectively, with mAP@0.5 = 0.985 and real-time performance of 64 FPS. The study validates that the proposed lightweight framework effectively balances accuracy and efficiency, and proves to be a practical solution for real-time defect detection in resource-constrained environments.
dc.language.isoEN
dc.publisherMDPI AG
dc.subject.lccIndustrial engineering. Management engineering
dc.titleOptimized Adaptive Multi-Scale Architecture for Surface Defect Recognition
dc.typeArticle
dc.description.keywordssteel surface defect detection
dc.description.keywordslightweight model
dc.description.keywordsrevcol
dc.description.keywordsgeneralization performance
dc.description.doi10.3390/a18080529
dc.title.journalAlgorithms
dc.identifier.e-issn1999-4893
dc.identifier.oaioai:doaj.org/journal:d159ad4771664bddb6325c5ad4729738
dc.journal.infoVolume 18, Issue 8


This item appears in the following Collection(s)

Show simple item record