Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

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

Allele-specific real-time PCR testing for minor HIV-1 drug resistance mutations: assay preparation and application to reveal dynamic of mutations in vivo.

Chinese medical journal·2011
Same author

Phase synchronization of foot dynamics in quiet standing.

Neuroscience letters·2011
Same author

Multiphase-computed tomography-based target volume definition in conventional fractionated radiotherapy of lung tumors: dosimetric and reliable comparison with the technique using addition of generic margins.

Tumori·2011
Same author

The effect of intracellular amino acids on GSH production by high-cell-density cultivation of Saccharomyces cerevisiae.

Applied biochemistry and biotechnology·2011
Same author

Nitrogen-doped graphene/ZnSe nanocomposites: hydrothermal synthesis and their enhanced electrochemical and photocatalytic activities.

ACS nano·2011
Same author

Dissection of genetic pathways in C. elegans.

Methods in cell biology·2011

Related Experiment Video

Updated: Jul 15, 2025

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography
11:34

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Published on: May 15, 2017

11.2K

FS-RSDD: Few-Shot Rail Surface Defect Detection with Prototype Learning.

Yongzhi Min1, Ziwei Wang1, Yang Liu1

  • 1School of Automation and Electrical Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China.

Sensors (Basel, Switzerland)
|September 28, 2023
PubMed
Summary

This study introduces FS-RSDD, a novel rail surface defect detection model. It effectively identifies rail damage using few-shot learning, enhancing railway safety with high accuracy.

Keywords:
few-shot learningprototype learningrail surface defect detectiontransfer learningunsupervised anomaly detection

More Related Videos

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
08:27

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

Published on: January 5, 2024

1.1K
Automated Rat Single-Pellet Reaching with 3-Dimensional Reconstruction of Paw and Digit Trajectories
07:52

Automated Rat Single-Pellet Reaching with 3-Dimensional Reconstruction of Paw and Digit Trajectories

Published on: July 10, 2019

14.2K

Related Experiment Videos

Last Updated: Jul 15, 2025

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography
11:34

Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography

Published on: May 15, 2017

11.2K
Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
08:27

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

Published on: January 5, 2024

1.1K
Automated Rat Single-Pellet Reaching with 3-Dimensional Reconstruction of Paw and Digit Trajectories
07:52

Automated Rat Single-Pellet Reaching with 3-Dimensional Reconstruction of Paw and Digit Trajectories

Published on: July 10, 2019

14.2K

Area of Science:

  • Railway Engineering
  • Computer Vision
  • Machine Learning

Background:

  • Rail surface damage poses significant safety risks in railway systems.
  • Existing defect detection models struggle with insufficient defect samples and validation credibility.

Purpose of the Study:

  • To propose FS-RSDD, a simple and effective model for rail surface defect detection.
  • To address the challenge of limited defect samples in current detection models.
  • To enhance rail surface condition monitoring and railway safety.

Main Methods:

  • Utilizes a pre-trained model for deep feature extraction from normal and defect rail samples.
  • Employs unsupervised learning to establish a feature prototype memory bank.
  • Applies prototype learning for pixel-wise defect probability estimation.

Main Results:

  • FS-RSDD achieves high accuracy in defect detection and localization with minimal defect samples.
  • Demonstrates superior performance compared to benchmarked few-shot industrial defect detection algorithms.
  • Achieves ROC scores of 95.2% and 99.1% on RSDDS Type-I and Type-II rail defect data, respectively.
  • Performs comparably to state-of-the-art unsupervised anomaly detection algorithms.

Conclusions:

  • FS-RSDD effectively overcomes limitations of supervised learning models with insufficient data.
  • The proposed model offers a robust solution for rail surface condition monitoring.
  • This approach enhances railway safety through accurate and efficient defect detection.