Related Experiment Video
Updated: Oct 13, 2025
08:27
Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
Published on: January 5, 2024
1.3K
Multivariate Analysis of Concrete Image Using Thermography and Edge Detection
Bubryur Kim1, Se-Woon Choi2, Gang Hu3
1Department of Robot and Smart System Engineering, Kyungpook National University, 80, Daehak-ro, Buk-gu, Daegu 41566, Korea.
Sensors (Basel, Switzerland)
|November 13, 2021
Summary
This study introduces an automated image analysis method for detecting structural damage like cracks. Utilizing neural networks and image segmentation, the system achieves 98% accuracy in classifying defects, enhancing structural health monitoring.
Area of Science:
- Engineering
- Computer Science
Background:
- Structural health monitoring (SHM) systems increasingly rely on data imaging for routine inspections.
- Image analysis of exterior damages provides crucial information on infrastructure integrity.
- Automated, reliable defect reporting in images is essential for efficient SHM.
Purpose of the Study:
- To present a multivariate image analysis approach for assessing structural damage, specifically cracks.
- To develop an automated system for robust and reliable defect detection in images.
- To enhance infrastructure maintenance through advanced image processing techniques.
Main Methods:
- Applied multivariate analysis and image processing techniques, including grayscale conversion.
- Implemented image segmentation for easier image transformation and analysis.
- Utilized a neural network for classifying visual characteristics of structural defects.
- Preprocessed concrete structure images to highlight cracks.
Main Results:
- Achieved 98% classification accuracy for identifying structural defects.
- Demonstrated that thermal image extraction yields superior histogram and cumulative distribution function features.
- The developed system effectively categorizes visual characteristics of damaged regions.
Conclusions:
- The proposed image analysis system offers a reliable method for automated structural damage assessment.
- Thermal imaging shows promise for improved feature extraction in defect detection.
- This approach can advance thermal image applications in nonphysical visual recognition and fault detection.
