Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
October 3, 2025SensorsOpen Access

Robust Industrial Surface Defect Detection Using Statistical Feature Extraction and Capsule Network Architectures

View Full Paper
Ask AI
Bookmark
Share

Authors

AMAzeddine MjahadARAlfredo Rosado-Muñoz

Discussion

Loading...

Member takes

Overview

This analysis reveals high precision and sensitivity in surface defect detection using machine learning and deep learning techniques.

Key Points

  • Capsule-based architectures achieved up to 100.0% precision for detecting normal class defects.
  • Random Forest and KNN showed high sensitivity and precision, both reaching about 99.4%.
  • ResNet50 attained 98.0% accuracy, indicating strong performance in analyzing surface defects.
  • Statistical feature extraction in combination with ML and DL techniques supports robust quality control in manufacturing.

Cite This Study

Mjahad et al. (2025) studied this question.

synapsesocial.com/papers/68e034f7f0e39f13e7fa3392https://doi.org/10.3390/s25196063
View Full Paper
Ask AI
Bookmark
Share