A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Image processing and Machine learning in Concrete Cube Crack detection
Concrete cube testing plays a crucial role in various aspects of modern construction. The structural performance of concrete cubes under direct compressive stress can result in failure through concrete cube breakout. Failure modes related to concrete can be classified into two types: acceptable and non-acceptable, with further classification into various modes. However, most of the time approximately 80% to 90% of the cubes are inaccurately selected, leading to lower strength and sustainability of concrete. Moreover, the excessive usage of cement required due to these inaccuracies contributes to global warming and increases costs. To address these issues, this research aims to develop an industry 4.0 solution for the construction and civil engineering fields. The proposed solution will be reliable, efficient, and based on image processing techniques. Convolutional Neural Networks (CNN) is used to detect and analyze cracks in concrete cubes. By examining the crack patterns, the damage area can be determined. By leveraging industry 4.0 technologies and advanced analysis techniques, this research aims to revolutionize the way concrete cube testing is conducted. The proposed solution will provide a reliable and efficient method for evaluating concrete cube quality, mitigating the negative impacts associated with inaccurate cube selection, and ultimately improving the performance and environmental sustainability of concrete in construction applications.
Image processing and Machine learning in Concrete Cube Crack detection
Concrete cube testing plays a crucial role in various aspects of modern construction. The structural performance of concrete cubes under direct compressive stress can result in failure through concrete cube breakout. Failure modes related to concrete can be classified into two types: acceptable and non-acceptable, with further classification into various modes. However, most of the time approximately 80% to 90% of the cubes are inaccurately selected, leading to lower strength and sustainability of concrete. Moreover, the excessive usage of cement required due to these inaccuracies contributes to global warming and increases costs. To address these issues, this research aims to develop an industry 4.0 solution for the construction and civil engineering fields. The proposed solution will be reliable, efficient, and based on image processing techniques. Convolutional Neural Networks (CNN) is used to detect and analyze cracks in concrete cubes. By examining the crack patterns, the damage area can be determined. By leveraging industry 4.0 technologies and advanced analysis techniques, this research aims to revolutionize the way concrete cube testing is conducted. The proposed solution will provide a reliable and efficient method for evaluating concrete cube quality, mitigating the negative impacts associated with inaccurate cube selection, and ultimately improving the performance and environmental sustainability of concrete in construction applications.
Image processing and Machine learning in Concrete Cube Crack detection
Patil, Meenakshi Somnath (author) / Ghongade, R.B. (author) / Salunke, Rupali Vilas (author)
2023-10-27
551500 byte
Conference paper
Electronic Resource
English
Review of Image Based Concrete Crack Detection
| IEEE | 2021
Concrete Cube-Shaped Display Hall
| British Library Conference Proceedings | 1992
Concrete Crack Detection by Multiple Sequential Image Filtering
| Online Contents | 2012