Technology of crack detection in reinforced concrete structures

• Main Authors: Wan ]usoh, Wan Amizah, Hanipah, Mohd Hafizal, Zakaria, Mohd Azuan, Pakir, Faizal, Adnan, Suraya Hani, Osman, Mohamad Hairi
• Other Authors: Tuan Ismail, Tuan Noor Hasanah
• Format: Book Section
• Language: English
• Published: Penerbit UTHM 2020
• Subjects: TH Building construction
• Online Access: http://eprints.uthm.edu.my/2216/1/Ch11%20Technology%20of%20Crack%20Detection.pdf; http://eprints.uthm.edu.my/2216/

Some crucial signs of structural failure that are critical for repair would be cracks on the structures as well as constant exposure that can result in severe environmental damage. Being able to detect cracks on structures is becoming an essential aspect of the technology of the construction industry. Destructive Testing and Non-Destructive Testing are the two methods used for structural crack detection. This study focused on the techniques used to detect cracks. Several effective methods to detect cracks were carried out and compared to identify the most suitable method in detecting cracks on structures within the demographics of Malaysia. Image processing techniques (IPTs) through the photogrammetry method, surface crack analysis program and Convolution Neural Network (CNN) were carried out to examine crack detection through measurement and monitoring from images. The distance was determined in this study for the physical properties, using both conductibility and accuracy. The photogrammetry method was able to conduct distance at 0.1 - 40 m, with an accuracy of up to 0.005 mm. Therefore, the surface cracks analysis provided 0.10 mm accuracy, while results on CNN had an accuracy of 0.95 mm (98.22 % and 97.95 % in training and validation). Results from physical properties showed that photogrammetry had the highest accuracy, while CNN has the least accuracy. Hence, this study concluded that Photogrammetry method and Convolution Neural Network (CNN) were both the most effective methods to be used in providing clear information and effective ways to detect crack on structures.