Related Concept Videos
Microcracking in Concrete
Types of Non-structural Cracks in Concrete
Plastic shrinkage cracks typically form within hours after the concrete is poured. The concrete's surface dries faster than the bottom, creating tensile stress that the still-plastic concrete cannot withstand, leading to diagonal or randomly patterned cracks on the concrete surface.
Non-destructive Tests for Concrete Strength
Design Example: Joints in Concrete Pavements
Contraction joints are typically formed by sawing a groove into the concrete shortly after it has hardened. This creates a weakened vertical plane, deliberately encouraging cracking at...
Rolling Resistance: Problem Solving
Elastic Collisions: Case Study
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Triad-LMF: a hierarchical low-rank multimodal fusion framework for robust cancer subtype classification using multi-omics data.
TeaWeeding-Action: a vision-based dataset for weeding behavior recognition in tea plantations.
Constitutive Model of Secondary Annealing Behavior of Cu-Cu Joints in Cu/SiO<sub>2</sub> Hybrid Bonding.
Related Experiment Video
Updated: Sep 19, 2025

Crack Monitoring in Resonance Fatigue Testing of Welded Specimens Using Digital Image Correlation
Published on: September 29, 2019
Flexi-YOLO: A lightweight method for road crack detection in complex environments.
Jiexiang Yang1, Renjie Tian2, Zexing Zhou1
1School of Information Science and Engineering, Chongqing Jiaotong University, Chongqing, China.
This study introduces Flexi-YOLO, a novel lightweight model for accurate road crack detection. It significantly improves detection accuracy and robustness, meeting industrial demands for real-time applications.
Area of Science:
- Computer Vision
- Artificial Intelligence
- Civil Engineering
Background:
- Road crack detection is vital for infrastructure maintenance and public safety.
- Complex backgrounds and crack patterns pose challenges for current detection methods.
- Existing models often struggle with real-time, efficient, and accurate detection.
Purpose of the Study:
- To propose a lightweight and robust model for efficient and accurate road crack detection.
- To enhance the performance of object detection algorithms for infrastructure monitoring.
- To address the limitations of current methods in handling complex crack features and environments.
Main Methods:
- Developed Flexi-YOLO, a lightweight model based on the YOLOv8 algorithm.
- Integrated Wise-IoU loss function for improved bounding box regression and sample robustness.
- Incorporated DCNv-C2f module for adaptive feature transformation and fusion.
- Utilized Global Attention Module (GAM) and AKConv for enhanced global and local feature perception.
- Implemented a lightweight G-Head (Ghost-Head) detection head to reduce feature redundancy.
Main Results:
- Flexi-YOLO achieved a 2.7% increase in accuracy and a 4.7% rise in recall over YOLOv8n.
- mAP improved by 5.3% and mAP@0.5-0.95 by 3.9%.
- Reduced GFLOPS by 0.5 and improved F1 score from 0.80 to 0.84.
- Demonstrated enhanced robustness to low-quality samples and complex crack patterns.
Conclusions:
- Flexi-YOLO provides a highly accurate and robust solution for automated road crack detection.
- The model's lightweight design meets industrial requirements for real-time processing and cost-effectiveness.
- This approach offers significant improvements for infrastructure maintenance and public safety applications.

