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

High-throughput characterization of transcription factors that modulate UV damage formation and repair at single-nucleotide resolution.

Nature communications·2026

Same author

Associations of unsweetened and sweetened coffee and tea consumption with incident cardiovascular disease: a large prospective cohort study.

European journal of nutrition·2026

Same author

Cryo-electron microscopy structures of human cone visual pigments.

Science (New York, N.Y.)·2026

Same author

Lignin-Regulated Amphiphilic Janus Membrane for Eco-Friendly Electronic Skin.

ACS sensors·2026

Same author

High-flow nasal oxygen therapy via a single-prong cannula interface during bronchoscopy in patients with acute respiratory failure: a two-center, open-label, randomized controlled trial.

Annals of intensive care·2026

Same author

Structural basis of the neuronal M-current generated by an asymmetric KCNQ2/3 assembly.

Cell research·2026

Search research articles

Related Experiment Video

Updated: Jul 9, 2025

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Published on: May 15, 2017

Defects detection of GMAW process based on convolutional neural network algorithm.

Haichao Li¹, Yixuan Ma², Mingrui Duan¹

¹State Key Laboratory of Advanced Welding and Joining, Harbin Institute of Technology, Harbin, 150001, China.

Scientific Reports

|December 1, 2023

Summary

This summary is machine-generated.

This study introduces a convolutional neural network (CNN) algorithm for predicting welding quality in gas metal arc welding. The CNN model achieves over 95% accuracy in detecting defects like penetration, craters, and slags using molten pool images.

More Related Videos

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

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

Related Experiment Videos

Last Updated: Jul 9, 2025

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Published on: May 15, 2017

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

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

Area of Science:

Materials Science
Manufacturing Engineering
Artificial Intelligence

Background:

Predicting welding quality is crucial for gas metal arc welding (GMAW) processes.
Existing methods may struggle with arc light interference, impacting molten pool image clarity.

Purpose of the Study:

To develop a robust welding defect detection algorithm using Convolutional Neural Networks (CNNs).
To accurately predict welding penetration, crater formation, and slag defects by analyzing molten pool images.
To enhance the algorithm's robustness across diverse welding scenarios.

Main Methods:

Developed a sensing system and image processing algorithm to capture clear molten pool images, overcoming arc light interference.
Utilized a CNN model trained and tested on a dataset of molten pool images.
Optimized CNN parameters (kernel-size, batch-size, learning rate) for improved prediction accuracy.

Main Results:

Achieved a prediction accuracy exceeding 95% for welding defects.
Successfully correlated molten pool visual features with specific defects such as burn-through (black hole), surface pores (circular voids), and fusion holes.
Demonstrated the model's ability to identify excessive penetration when the crater lacks concavity.

Conclusions:

The developed CNN algorithm effectively predicts welding quality and detects defects in GMAW.
The model's analysis of molten pool characteristics provides insights into defect formation mechanisms.
The algorithm shows enhanced robustness and potential for real-world welding applications.