• Home
  • Search
  • LHO-net: A Lightweight Steel Defect Detection Framework Based on Cross-Scale Feature Selection and Adaptive Optimization.
  • https://doi.org/10.3390/s26061990Copy DOI Icon

LHO-net: A Lightweight Steel Defect Detection Framework Based on Cross-Scale Feature Selection and Adaptive Optimization.

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

To address the issues of poor adaptability to complex scenarios, high computational complexity, and difficulties in terminal deployment of existing steel surface defect detection models, a novel lightweight detection network named LHO-net is proposed, with the Lightweight Multi-Backbone (LM Backbone), the Hierarchical Scale-based Pyramid Attention Network (HSPAN), and the Occlusion-aware Detection Head (OAHead). The LM Backbone adopts a dual-branch structure with shared HGStem and a dynamic feature fusion mechanism, effectively capturing multi-dimensional features of irregular defects while extremely compressing model parameters. The HSPAN module realizes efficient fusion of multi-scale features through dynamic feature selection and adaptive upsampling strategies, balancing background noise suppression and defect detail preservation. The OAHead completes adaptive compensation of features in occluded regions by means of deep feature aggregation and exponential normalization technology, significantly enhancing the ability to recognize complex defects. On the NEU-DET dataset, LHO-net achieves a mAP@0.5 of 75.0%, a mAP@0.5:0.95 of 44.0%, and a recall of 73.6%, with a computational complexity of only 2.3 GFLOPS. Compared with the baseline model YOLOv12, it reduces parameters by 64% and computational cost by 60.3%. On the GC-10 dataset, its mAP@0.5 reaches 67.2%, and its detection stability for complex defects such as slender creases and low-contrast water spots is superior to that of mainstream lightweight YOLO variants. Visualization results confirm that the model can effectively avoid common problems such as redundant annotations and false detections and maintains stable recognition performance for various defects. It solves the core contradiction between detection accuracy and lightweight deployment in industrial scenarios, providing an efficient and practical technical solution for real-time steel surface defect detection on resource-constrained terminal devices.

Similar Papers
  • Research Article
  • Citations2

A Global Feature Reused Network for Defect Detection in Steel Images

  • Sep 09, 2024
  • Journal of Computing and Information Science in Engineering
  • Chengli Yang +3
  • PDF
  • Research Article
  • Citations9

The Steel Surface Multiple Defect Detection and Size Measurement System Based on Improved YOLOv5

  • May 20, 2023
  • Journal of Electrical and Computer Engineering
  • Yiming Xu +4
  • Research Article

Faster RCNN based on Feature Fusion and its Application in Steel Surface Defect Detection

  • Oct 23, 2022
  • Scientific Journal of Technology
  • Shiwei Xu +1
  • PDF
  • Research Article
  • Citations10

Steel surface defect detection method based on improved YOLOv9

  • Jul 11, 2025
  • Scientific Reports
  • Cong Chen +2
  • Research Article

Surface Defect Detection of Steel Based on Improved YOLOv5s

  • Aug 27, 2025
  • International Journal of Computer Science and Information Technology
  • Ziyan Sha +3
  • Research Article
  • Citations7

Steel Surface Defect Detection Technology Based on YOLOv8-MGVS

  • Jan 23, 2025
  • Metals
  • Kai Zeng +5
  • PDF
  • Research Article
  • Citations88

Real-Time Steel Surface Defect Detection with Improved Multi-Scale YOLO-v5

  • Apr 28, 2023
  • Processes
  • Ling Wang +4
  • Research Article
  • Citations1

A Review of Deep Learning-Based Steel Surface Defect Detection

  • Apr 21, 2025
  • Academic Journal of Science and Technology
  • Feiyu Chen +5
  • Conference Article
  • Citations2

Pattern classification of deep brain local field potentials for brain computer interfaces

  • Dec 01, 2012
  • K A Mamun +8
  • Research Article

Confidence-gradient reweighting and lightweight feature enhancement algorithm for steel surface defect detection.

  • Jan 18, 2026
  • Scientific reports
  • Linxuan Chen +5
  • PDF
  • Research Article
  • Citations6

Research on Steel Surface Defect Detection Algorithm Based on Improved Deep Learning

  • Dec 30, 2022
  • International Journal of Electrical and Electronics Research
  • Fei Ren +4
  • Supplementary Content

CLIP-Driven with Dynamic Feature Selection and Alignment Network for Referring Remote Sensing Image Segmentation

  • Sep 17, 2025
  • Preprints.org
  • Qianqi Lu +6
  • Research Article
  • Citations1

A lightweight multi-wedge belt defect detection method combining CNN and transformer

  • Jan 01, 2026
  • Engineering Research Express
  • Junyi Wang +7
  • Book Chapter
  • Citations7

Dynamic Feature Selection Algorithm Based on Minimum Vertex Cover of Hypergraph

  • Jan 01, 2018
  • Xiaojun Xie +1
  • Research Article
  • Citations95

Selection of Dynamic Features Based on Time–Frequency Representations for Heart Murmur Detection from Phonocardiographic Signals

  • Nov 17, 2009
  • Annals of Biomedical Engineering
  • A F Quiceno-Manrique +3
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.