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

A High-Precision Method for Extracting Lateral Deformation in Operational Shield Tunnels Based on LiDAR Point Cloud Analysis.

Sensors (Basel, Switzerland)·2026
Same author

Mercury supply limits methylmercury production in paddy soils.

The Science of the total environment·2024
Same author

Soil Geobacteraceae are the key predictors of neurotoxic methylmercury bioaccumulation in rice.

Nature food·2024
Same author

Improving the Accuracy of an R-CNN-Based Crack Identification System Using Different Preprocessing Algorithms.

Sensors (Basel, Switzerland)·2022
Same author

Dynamic Response of CFRP Reinforced Steel Beams Subjected to Impact Action Based on FBG Sensing Technology.

Sensors (Basel, Switzerland)·2022
Same author

Safety Risk Analysis of a New Design of Basalt Fiber Gabion Slope Based on Improved 3D Discrete Element Method and Monitoring Data.

Sensors (Basel, Switzerland)·2022

Related Experiment Video

Updated: Jul 16, 2026

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

CAMD-RTDETR: Real-Time Multi-Defect Detection Method for Tunnel Structures.

Yunyun Hao1, Xiangyang Xu1

  • 1School of Rail Transportation, Soochow University, Suzhou 215006, China.

Sensors (Basel, Switzerland)
|July 15, 2026
PubMed
Summary

This study introduces CAMD-RTDETR, a new deep learning model for real-time tunnel defect detection. It significantly improves accuracy and speed for identifying cracks, seepage, and spalling in tunnels.

Keywords:
CAMD-RTDETRcross-attention feature miningdecoding enhancementdeep learningmulti-scale feature poolingreal-time object detectiontunnel multi-defect detection

More Related Videos

A Basic Positron Emission Tomography System Constructed to Locate a Radioactive Source in a Bi-dimensional Space
14:19

A Basic Positron Emission Tomography System Constructed to Locate a Radioactive Source in a Bi-dimensional Space

Published on: February 1, 2016

Computed Tomography-guided Time-domain Diffuse Fluorescence Tomography in Small Animals for Localization of Cancer Biomarkers
12:24

Computed Tomography-guided Time-domain Diffuse Fluorescence Tomography in Small Animals for Localization of Cancer Biomarkers

Published on: July 17, 2012

Related Experiment Videos

Last Updated: Jul 16, 2026

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

A Basic Positron Emission Tomography System Constructed to Locate a Radioactive Source in a Bi-dimensional Space
14:19

A Basic Positron Emission Tomography System Constructed to Locate a Radioactive Source in a Bi-dimensional Space

Published on: February 1, 2016

Computed Tomography-guided Time-domain Diffuse Fluorescence Tomography in Small Animals for Localization of Cancer Biomarkers
12:24

Computed Tomography-guided Time-domain Diffuse Fluorescence Tomography in Small Animals for Localization of Cancer Biomarkers

Published on: July 17, 2012

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Structural Engineering

Background:

  • Manual tunnel inspection is inefficient and risky.
  • Existing deep learning methods struggle with small targets, scale variations, and background noise in tunnel defect detection.
  • Real-time deployment on edge devices is limited by current methods' performance.

Purpose of the Study:

  • To develop an end-to-end, real-time multi-defect detection method for tunnels.
  • To enhance the accuracy and stability of tunnel defect detection in complex environments.
  • To enable efficient deployment on edge devices for structural safety and maintenance.

Main Methods:

  • Proposed CAMD-RTDETR, an end-to-end real-time multi-defect detection method based on RT-DETR.
  • Introduced cross-attention feature mining for improved perception of weak-texture defects.
  • Implemented multi-scale contextual pooling for unified representation of diverse defects.
  • Incorporated decoding enhancement and query optimization for better detection stability and boundary accuracy.

Main Results:

  • CAMD-RTDETR achieved an average inference latency of 15.855 ms/image and 63.06 FPS.
  • Demonstrated significant improvements over baseline RT-DETR: Precision (+6.3%), Recall (+13.5%), mAP50 (+14.7%), mAP50-95 (+15.8%).
  • Outperformed seven other representative detectors in accuracy and real-time performance.

Conclusions:

  • CAMD-RTDETR shows superior accuracy and real-time capabilities for tunnel defect detection.
  • The method is feasible for edge-side inference in tunnel inspection systems.
  • Potential for integration into vehicle-mounted systems for enhanced structural monitoring.