Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
August 11, 2025Applied and Computational Engineering

Evaluating the Detection Performance of Different Convolutional Neural Network Models for Steel Surface Defects

View Full Paper
Ask AI
Bookmark
Share

Authors

ZCZhichao Chen

Discussion

Loading...

Member takes

Overview

Performance evaluation shows VGG16 achieved 99.26% accuracy in detecting steel surface defects, suggesting optimal model selection criteria.

Key Points

  • VGG16 achieved the highest accuracy at 99.26%, indicating superior performance among the models evaluated.
  • Models tested include VGG16, ResNet50, InceptionV3, and MobileNetV2, all demonstrating over 97% accuracy in classification.
  • Unified data preprocessing and augmentation strategies were adopted across all models to ensure consistent evaluation.
  • The results highlight that MobileNetV2 offers a balance of high classification accuracy and computational efficiency.

Cite This Study

Zhichao Chen (2025) studied this question.

synapsesocial.com/papers/68a360e70a429f79733298bahttps://doi.org/10.54254/2755-2721/2025.bj25449
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. 1Performance evaluation of CNN models for steel surface defect detection in lean manufacturing2025 · 3 citations
  2. 2Learning-Based Approach for Automated Surface Inspection with Industrial Tomography Imaging2024 · 7 citations
  3. 3Performance Evaluation and Misclassification Distribution Analysis of Pre-Trained Lightweight CNN Models for Hot-Rolled Steel Strip Surface Defect Classification Under Degraded Imaging Conditions2025
  4. 4Model-Centric or Data-Centric Approach? A Case Study on the Classification of Surface Defects in Steel Hot Rolling Using Convolutional Neural Networks2026 · 1 citations
  5. 5Cross-Dataset Benchmarking of Deep Learning Models for Surface Defect Classification in Metal Parts2026 · 2 citations