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

Fabrication of Terahertz Fresnel Zone Plates via Ultraprecision Mechanical Processing.

Micromachines·2026
Same author

Numerical Study of Quantifying Diffusion Effects on Deterministic Lateral Displacement at the Nanoscale.

Analytical chemistry·2026
Same author

Design and fabrication of an aluminium oxide cutting insert with an internal cooling channel.

Advances in manufacturing·2024
Same author

First-principles study of electron dynamics of MoS2 under femtosecond laser irradiation from deep ultraviolet to near-infrared wavelengths.

The Journal of chemical physics·2024
Same author

Stereo Bi-Telecentric Phase-Measuring Deflectometry.

Sensors (Basel, Switzerland)·2024
Same author

Study on magnetohydrodynamic internal cooling mechanism within an aluminium oxide cutting tool.

The International journal, advanced manufacturing technology·2024

Related Experiment Video

Updated: Dec 28, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

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

Published on: August 30, 2013

43.5K

Automatic Fabric Defect Detection Using Cascaded Mixed Feature Pyramid with Guided Localization.

You Wu1, Xiaodong Zhang1, And Fengzhou Fang1

  • 1State Key Laboratory of Precision Measuring Technology & Instruments, Centre of Micro/Nano Manufacturing Technology-MNMT, Tianjin University, Tianjin 300072, China.

Sensors (Basel, Switzerland)
|February 12, 2020
PubMed
Summary
This summary is machine-generated.

This study introduces an improved fabric defect detection system for optical images, addressing challenges like multi-scale and blurred defects. The new detector enhances performance, particularly for occluded fabric flaws.

Keywords:
cascaded center-nesscross-scaledeformable localizationfabric defectmixed kernelsobject detection

More Related Videos

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

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

Published on: December 15, 2023

941
Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

11.1K

Related Experiment Videos

Last Updated: Dec 28, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

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

Published on: August 30, 2013

43.5K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

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

Published on: December 15, 2023

941
Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

11.1K

Area of Science:

  • Computer Vision
  • Materials Science
  • Industrial Automation

Background:

  • Generic object detection algorithms perform well on natural images but struggle with fabric defect datasets.
  • Fabric defect images present unique challenges: multi-scale, noise, blur, and sensitivity to backlight intensity.
  • Existing methods lack the generalization needed for practical industrial fabric defect detection.

Purpose of the Study:

  • To systematically study and improve fabric defect detection on optical image datasets.
  • To develop a robust, generalized object detection system specifically for industrial fabric inspection.
  • To enhance the detection of challenging fabric defects, including occluded ones.

Main Methods:

  • Collection and curation of large-scale, imbalanced fabric defect datasets.
  • Construction of an improved two-stage defect detector with stacked feature pyramid networks.
  • Integration of interpolating mixed depth-wise blocks and deformable convolutions for feature aggregation and refinement.
  • Utilizing balanced sampling and position-sensitive pooling for region of interest characterization.

Main Results:

  • The proposed two-stage detector demonstrates improved generalization capabilities for fabric defect detection.
  • Stacked feature pyramid networks effectively aggregate cross-scale defect patterns.
  • The architecture shows superior performance in detecting occluded defects compared to existing region-based object detectors.
  • The system addresses the imbalanced nature and specific challenges of fabric defect datasets.

Conclusions:

  • The developed end-to-end architecture significantly advances fabric defect detection accuracy and robustness.
  • This improved detector is suitable for practical industrial applications requiring high-precision fabric inspection.
  • The study highlights the importance of tailored algorithms for specialized image datasets in object detection.