Concrete Subsurface Crack Detection Using Thermal Imaging in a Deep Neural Network
| Main Author: | Abuhmida, Mabrouka |
|---|---|
| Format: | Article info application/pdf eJournal |
| Bahasa: | eng |
| Terbitan: |
IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia
, 2023
|
| Subjects: | |
| Online Access: |
https://journal.ugm.ac.id/ijccs/article/view/82912 https://journal.ugm.ac.id/ijccs/article/view/82912/36326 |
Daftar Isi:
- The article discusses how impact actions, such as conflict and warfare, can negatively impact the structural integrity of concrete structures and how detecting hidden defects in concrete structures is difficult without expert knowledge. The paper presents a new technique that combines thermal imaging and artificial intelligence to detect hidden defects in concrete structures. The authors trained an AI model on simulated data and achieved a validation accuracy of 99.93%. They then conducted a laboratory experiment to create a dataset of concrete blocks with and without subsurface cracks and trained a new model, which achieved a validation accuracy of 100%. The article concludes that AI can detect hidden defects and subsurface cracks in concrete structures by classifying thermal images of concrete surfaces.