Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Optimizing Tumor Detection in Brain MRI with One-Class SVM and Convolutional Neural Network-Based Feature Extraction.

Journal of imaging·2025

Same author

Efficient Extraction of Deep Image Features Using a Convolutional Neural Network (CNN) for Detecting Ventricular Fibrillation and Tachycardia.

Journal of imaging·2023

Same author

A Smart Agricultural System Based on PLC and a Cloud Computing Web Application Using LoRa and LoRaWan.

Sensors (Basel, Switzerland)·2023

Same author

A Novel Approach of a Low-Cost UWB Microwave Imaging System with High Resolution Based on SAR and a New Fast Reconstruction Algorithm for Early-Stage Breast Cancer Detection.

Journal of imaging·2022

Same author

Food Tray Sealing Fault Detection in Multi-Spectral Images Using Data Fusion and Deep Learning Techniques.

Journal of imaging·2021

Same author

Sensor Node Network for Remote Moisture Measurement in Timber Based on Bluetooth Low Energy and Web-Based Monitoring System.

Sensors (Basel, Switzerland)·2021

Search research articles

Related Experiment Video

Updated: Jan 15, 2026

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Published on: January 5, 2024

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

Azeddine Mjahad¹, Alfredo Rosado-Muñoz¹

¹GDDP, Department Electronic Engineering, School of Engineering, University of Valencia, 46100 Burjassot, Valencia, Spain.

Sensors (Basel, Switzerland)

|October 16, 2025

Summary

This summary is machine-generated.

This study demonstrates that combining machine learning (ML) and deep learning (DL) with statistical features offers highly accurate automated surface defect detection for metallic castings. The developed methods achieve excellent precision and sensitivity, supporting real-time industrial applications.

Keywords:

CNN3D Industry 4.0 ResNet50 automated visual inspection capsule networks casting manufacturing defect detection feature selection

More Related Videos

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Simulation of a Scaled Assembly Process with Collaboration of a Robotic Arm and Monitoring through a Vision System for Quality Control

Simulation of a Scaled Assembly Process with Collaboration of a Robotic Arm and Monitoring through a Vision System for Quality Control

Published on: August 29, 2025

Related Experiment Videos

Last Updated: Jan 15, 2026

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Published on: January 5, 2024

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Simulation of a Scaled Assembly Process with Collaboration of a Robotic Arm and Monitoring through a Vision System for Quality Control

Simulation of a Scaled Assembly Process with Collaboration of a Robotic Arm and Monitoring through a Vision System for Quality Control

Published on: August 29, 2025

Area of Science:

Materials Science and Engineering
Computer Science
Artificial Intelligence

Background:

Automated quality control is essential for manufacturing metallic cast components.
Fast and accurate surface defect detection is a key challenge in this domain.
Traditional methods often lack the speed and precision required for modern industrial demands.

Purpose of the Study:

To evaluate classical Machine Learning (ML) algorithms and deep learning (DL) architectures for automated surface defect detection in metallic cast components.
To compare the performance of ML models using statistical parameters against DL models using image inputs.
To assess the potential for real-time application through processing time measurements.

Main Methods:

Evaluated ML algorithms (Random Forest, Gradient Boosting, K-Nearest Neighbors, SVM) with extracted statistical parameters.
Assessed DL architectures including ResNet50, Capsule Networks (ConvCapsuleLayer), and a 3D Convolutional Neural Network (CNN3D).
Utilized both original and expanded datasets, employing repeated train-test splits for robust metric calculation (accuracy, precision, recall, F1-score).

Main Results:

ML models like Random Forest achieved high performance (e.g., 99.4% precision and sensitivity) on the original dataset.
Capsule-based architectures demonstrated superior results, with ConvCapsuleLayer reaching 98.7% accuracy and 100% precision for the normal class.
All evaluated models exhibited very low per-image processing times, indicating suitability for real-time applications.

Conclusions:

Combining statistical descriptors with ML and DL architectures provides a robust and scalable solution for automated, non-destructive surface defect detection.
The evaluated methods demonstrate high accuracy and reliability across different datasets.
The findings support the implementation of these advanced techniques for efficient industrial quality control.