• Home
  • Search
  • Steel Surface Defect Segmentation Using U-Net with SLMA-Gray Attention and AdamW Optimization
  • Cite Icon1
  • https://doi.org/10.31449/inf.v49i31.10673Copy DOI Icon

Steel Surface Defect Segmentation Using U-Net with SLMA-Gray Attention and AdamW Optimization

Show More
  • Abstract
  • Literature Map
  • Citations
  • Similar Papers
Abstract

Rust detection on metal surfaces is a challenging task in industrial maintenance and quality control. This paper proposes an improved U-Net segmentation method that enhances the accuracy of surface defect detection in steel materials. This paper proposes an improved U-Net model for steel surface defect segmentation. We introduce a novel Spatial-Channel Gray-Level Mixed Attention mechanism (SLMA-Gray) to enhance defect saliency in grayscale images and employ the AdamW optimizer to improve generalization. Experiments on the NEU-DET dataset show our method significantly outperforms the original U-Net, with increases of 5.29% in precision, 18.84% in recall, 3.34% in accuracy, 15.17% in Dice, and 18.18% in IoU.

Similar Papers
  • Conference Article

Defects detection of high precision sieve based on improved YOLOv4

  • Sep 08, 2025
  • Kewei Liu +5
  • Research Article
  • Citations52

A multi-task and multi-scale convolutional neural network for automatic recognition of woven fabric pattern

  • Jun 26, 2020
  • Journal of Intelligent Manufacturing
  • Shuo Meng +5
  • Research Article

A Novel Analytical Method for Rapid Detection of Batch‐To‐Batch Variation and Storage Stability of Natural Flavouring Materials by Online Py‐ GC / MS Technology

  • Mar 02, 2026
  • Flavour and Fragrance Journal
  • Changrong Luo +8
  • Research Article

Application of quality control in product supply using deep learning and machine learning

  • Apr 07, 2026
  • Journal of Applied Microeconometrics
  • Murat Atan +2
  • Research Article
  • Citations2

Detection and classification of metal surface defects using lite convolutional neural network (LCNN)

  • Jul 03, 2024
  • Material Science & Engineering International Journal
  • Al-Mahmud Al Mamun +2
  • Research Article
  • Citations2

A lightweight image-level segmentation method for steel surface defects based on cross-layer feature fusion

  • Mar 01, 2024
  • Insight - Non-Destructive Testing and Condition Monitoring
  • Peng Wang +4
  • Research Article
  • Citations287

Triplet-Graph Reasoning Network for Few-Shot Metal Generic Surface Defect Segmentation

  • Jan 01, 2021
  • IEEE Transactions on Instrumentation and Measurement
  • Yanqi Bao +6
  • Research Article
  • Citations11

Real-Time Milk Quality Control Using Multi-Spectral Sensing and Edge Computing: Advancing On-Site Detection of Milk Components with XGBoost

  • Nov 25, 2024
  • Applied Sciences
  • Mahmut Durgun
  • Research Article
  • Citations40

Enhanced Circuit Board Analysis: Infrared Image Segmentation Utilizing Markov Random Field (MRF) and Level Set Techniques

  • Mar 01, 2025
  • Engineering Reports
  • T Praveenkumar +7
  • Research Article
  • Citations1

Surface defect detection of cylinder liner based on improved YOLOv5

  • Mar 23, 2024
  • Journal of Intelligent & Fuzzy Systems
  • Dai Jun +4
  • Research Article
  • Citations2

Design and Study of Pulsed Eddy Current Sensor for Detecting Surface Defects in Small-Diameter Bars.

  • Dec 18, 2024
  • Sensors (Basel, Switzerland)
  • Lei Han +2
  • Research Article

SAFETY OF CONSUMER GOODS AS THE MAIN PROBLEM OF MODERN COMMODITY SCIENCE

  • Jan 01, 2022
  • Herald of Khmelnytskyi National University
  • Viktoriia Vasiuta +2
  • Research Article
  • Citations1

Machine learning for enhancing manufacturing quality control in ultrasonic nondestructive testing: A Wavelet Neural Network and Genetic Algorithm approach

  • Oct 31, 2024
  • Advances in Production Engineering & Management
  • W.T Song +1
  • Research Article
  • Citations75

The legume inoculant industry and inoculant quality control in Australia: 1953–2003

  • Apr 14, 2005
  • Australian Journal of Experimental Agriculture
  • G K Bullard +2
  • Research Article

Defect detection in polymer components using infrared thermography by employment of signal processing methods and U-Net semantic segmentation

  • Jul 16, 2025
  • Nondestructive Testing and Evaluation
  • Seyedmobin Mosavifard +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.