Related Experiment Video
Updated: May 21, 2025

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
Published on: January 5, 2024
An Optimized YOLOv11 Framework for the Efficient Multi-Category Defect Detection of Concrete Surface
Zhuang Tian1, Fan Yang1,2, Lei Yang1
1School of Transportation and Civil Engineering, Nantong University, Nantong 226019, China.
A new concrete defect detection model, YOLOv11-EMC, improves accuracy and reduces errors in identifying structural flaws. This enhances structural safety and extends the service life of concrete infrastructure.
Area of Science:
- Civil Engineering
- Computer Vision
- Artificial Intelligence
Background:
- Accurate concrete defect identification is vital for structural safety and longevity.
- Existing models struggle with the complexity and variability of concrete defects, leading to detection inaccuracies.
- These limitations pose risks to structural integrity in real-world engineering applications.
Purpose of the Study:
- To develop an efficient and robust concrete defect detection model.
- To enhance the accuracy and generalization capabilities of existing detection models.
- To address the challenges posed by irregular defect shapes and complexities in engineering inspections.
Main Methods:
- Proposed YOLOv11-EMC (efficient multi-category concrete defect detection) model.
- Incorporated modified deformable convolution to extract irregular defect features effectively.
- Integrated C3k2 module with a revised dynamic convolution module for computational efficiency and improved feature representation.
Main Results:
- YOLOv11-EMC demonstrated significant improvements over Yolov11 in precision (+8.3%), recall (+2.1%), mAP50 (+4.3%), and F1 (+3%).
- Field tests using drone imagery confirmed reduced false and under-detections.
- The model achieved higher overall detection accuracy in practical engineering scenarios.
Conclusions:
- YOLOv11-EMC offers a superior methodology for identifying tangible flaws in concrete structures.
- The enhanced model improves robustness and generalization for concrete defect detection.
- This contributes to better structural safety and extended service life in engineering applications.
Related Concept Videos
Non-destructive Tests for Concrete Strength
Workability of Concrete
Concrete's workability is determined by its resistance to internal forces that arise...
Design Example: Managing Concrete Workability
Abrasion Resistance of Concrete
One such test is the revolving disc test, where three plates...
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.
Reinforcements in Concrete

