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 Experiment Video

Updated: Jul 6, 2026

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments
03:31

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

Published on: December 15, 2023

Boundary-Sensitive Hybrid Attention Network for Multi-Scale Crack Fine Segmentation.

Yaotong Jiang1, Tianmiao Wang1, Congyu Shao1

  • 1School of Mechanical Engineering and Automation, Beihang University, Beijing 100191, China.

Sensors (Basel, Switzerland)
|May 27, 2026
PubMed
Summary
This summary is machine-generated.

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

Near-Infrared and Shortwave Infrared Building Blocks for Activity-Based Sensors in Animals.

Bioconjugate chemistry·2026
Same author

Real-world effect of vitrectomy combined with intravitreal conbercept for treatment of vitreous haemorrhage in proliferative diabetic retinopathy (QILIN I): a prospective cohort study.

The British journal of ophthalmology·2026
Same author

Research on Synthetic Data Methods and Detection Models for Micro-Cracks.

Sensors (Basel, Switzerland)·2026
Same author

Efficacy and safety of conbercept, ranibizumab and laser in retinopathy of prematurity: A systematic review and network meta-analysis.

Survey of ophthalmology·2026
Same author

Intensified anthropogenic activities dominate the spatiotemporal dynamics of antibiotic resistance genes in lake ecosystems.

Environmental research·2026
Same author

Vagus nerve stimulation enhances memory and long-term potentiation via the hippocampal NE/β-AR signaling pathway in pilocarpine rats.

Brain research bulletin·2026

This study introduces BSA-Net, a novel network for concrete crack segmentation in bridge health monitoring. It improves accuracy in challenging conditions, offering a reliable solution for automated infrastructure inspection.

Area of Science:

  • Civil Engineering
  • Computer Vision
  • Artificial Intelligence

Background:

  • Concrete crack segmentation is vital for infrastructure safety and longevity.
  • Traditional methods struggle with weak contrast, background noise, and multi-scale cracks.

Purpose of the Study:

  • To develop an advanced deep learning model for accurate concrete crack segmentation.
  • To address limitations of existing methods in complex and noisy environments.

Main Methods:

  • Introduced the Boundary-Sensitive Hybrid Attention Network (BSA-Net).
  • Employed a hierarchical Transformer encoder (Hiera-A) for multi-scale feature extraction.
  • Utilized a multi-scale context module (Light-ASPP) for efficient context aggregation.
  • Implemented a dual-branch boundary-aware decoder (BAD) for precise boundary detection.
Keywords:
Transformer encoderboundary-aware decoderbridge health monitoringcrack segmentationmulti-scale features

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Related Experiment Videos

Last Updated: Jul 6, 2026

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments
03:31

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

Published on: December 15, 2023

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Main Results:

  • BSA-Net demonstrated superior performance over existing crack detection models.
  • Achieved high accuracy in segmentation, boundary clarity, and recall rates, especially for subtle cracks.
  • Showcased effectiveness in complex, noisy environments and on benchmark datasets.

Conclusions:

  • BSA-Net offers a scalable and reliable solution for automated infrastructure monitoring.
  • The model enhances crack segmentation performance in real-world conditions.
  • Provides improved defect detection capabilities for civil infrastructure.