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

ProCoS: Degradation-Robust Manipulation Localization for Special Equipment Inspection Images.

Sensors (Basel, Switzerland)·2026

Same author

Study on Novel Surface Defect Detection Methods for Aeroengine Turbine Blades Based on the LFD-YOLO Framework.

Sensors (Basel, Switzerland)·2025

Same author

An Improved Generative Adversarial Network-Based and U-Shaped Transformer Method for Glass Curtain Crack Deblurring Using UAVs.

Sensors (Basel, Switzerland)·2024

Same author

Fast semantic segmentation method for machine vision inspection based on a fewer-parameters atrous convolution neural network.

PloS one·2021

Search research articles

Related Experiment Video

Updated: Sep 24, 2025

Visualization of Flow Field Around a Vibrating Pipeline Within an Equilibrium Scour Hole

Visualization of Flow Field Around a Vibrating Pipeline Within an Equilibrium Scour Hole

Published on: August 26, 2019

Pressure vessel-oriented visual inspection method based on deep learning.

Pu Liao¹, Liu Guixiong¹

¹School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou, China.

|May 2, 2022

Summary

This study introduces a vision and structured light method for precise pressure vessel weld inspection, improving efficiency and accuracy over manual methods. The new deep learning approach achieves theoretical measurement accuracy within 0.065 mm.

More Related Videos

Particle Image Velocimetry Investigation of Hemodynamics via Aortic Phantom

Particle Image Velocimetry Investigation of Hemodynamics via Aortic Phantom

Published on: February 25, 2022

Development and Evaluation of 3D-Printed Cardiovascular Phantoms for Interventional Planning and Training

Development and Evaluation of 3D-Printed Cardiovascular Phantoms for Interventional Planning and Training

Published on: January 18, 2021

Related Experiment Videos

Last Updated: Sep 24, 2025

Visualization of Flow Field Around a Vibrating Pipeline Within an Equilibrium Scour Hole

Visualization of Flow Field Around a Vibrating Pipeline Within an Equilibrium Scour Hole

Published on: August 26, 2019

Particle Image Velocimetry Investigation of Hemodynamics via Aortic Phantom

Particle Image Velocimetry Investigation of Hemodynamics via Aortic Phantom

Published on: February 25, 2022

Development and Evaluation of 3D-Printed Cardiovascular Phantoms for Interventional Planning and Training

Development and Evaluation of 3D-Printed Cardiovascular Phantoms for Interventional Planning and Training

Published on: January 18, 2021

Area of Science:

Mechanical Engineering
Computer Vision
Non-Destructive Testing

Background:

Traditional manual inspection of pressure vessel welds is inefficient and lacks accuracy.
Automated methods are needed to guarantee the safe operation of pressure vessels.

Purpose of the Study:

To develop an automated method for detecting surface parameters of pressure vessel welds.
To improve the efficiency and accuracy of weld inspection using computer vision and structured light.

Main Methods:

A deep convolution-based feature point extraction algorithm was developed for weld parameters.
A training data enhancement method using third-order non-uniform rational B-spline (NURBS) curves was employed.
A pressure vessel measurement device was designed and tested.

Main Results:

The proposed deep network demonstrated superior feature point extraction performance compared to DeepLabCut and HR-net.
The theoretical accuracy of surface parameter measurements was determined to be within 0.065 mm.

Conclusions:

The combined vision and structured light method offers a highly accurate and efficient solution for pressure vessel weld inspection.
The developed deep learning approach significantly enhances the precision of weld surface parameter measurement.