Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
February 6, 2026SensorsOpen Access

MobileSteelNet: A Lightweight Steel Surface Defect Classification Network with Cross-Interactive Efficient Multi-Scale Attention

View Full Paper
Ask AI
Bookmark
Share

Authors

XZXiang ZouJiangxi Normal UniversityZLZhongming LiuJiangxi Normal UniversityCXChengjun XuSouthwest University

Discussion

Loading...

Member takes

Implication

MobileSteelNet demonstrates enhanced steel surface defect classification, indicating significant improvements in accuracy and efficiency for industrial applications.

Key Points

  • The study aims to develop an efficient and accurate deep learning model for classifying steel surface defects in industrial settings.
  • Introduced MobileSteelNet with two novel modules: multi-scale feature fusion and Cross-Interactive Efficient Multi-Scale Attention.
  • Conducted experiments on the NEU-DET dataset to evaluate performance.
  • Compared results against existing models including ResNet-50 and MobileNetV2.
  • Achieved 91.36% average accuracy, outperforming ResNet-50's 88.01% and MobileNetV2's 86.08%.
  • Attained 93.70% accuracy specifically for Scratch-type defects.
  • Model size is 8.2 MB, making it suitable for edge deployment in real-time systems.

Cite This Study

Zou et al. (2026) studied this question.

synapsesocial.com/papers/698585db8f7c464f23009962https://doi.org/10.3390/s26031022
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1SCFNet: Lightweight Steel Defect Detection Network Based on Spatial Channel Reorganization and Weighted Jump Fusion2024 · 4 citations
  2. 2ID-MSNet: An Enhanced Multi-Scale Network with Convolutional Attention for Pixel-Level Steel Defect Segmentation2026 · 2 citations
  3. 3CMH‑Net:a structured and optimized network for real-time steel surface defect detection2025
  4. 4Rethinking multi-stage calibration fusion network and loss function for steel surface defect detection2025 · 1 citations
  5. 5Steel surface defect detection based on dynamic receptive field and multi-scale features fusion2026