Concrete Crack Detection Using Thermal Imaging and Neural Network

In the context of integrity testing for buildings, impact actions, including conflict and warfare, can have detrimental effects on the structural integrity of concrete structures. Even indirect impacts can lead to subsurface defects that compromise the safety of the structure. However, assessing these hidden defects often requires significant time and expert knowledge. Currently, there is a lack of techniques available for rapid assessment of usability and safety without expert intervention. This study proposes a novel approach combining thermal imaging and artificial intelligence (AI) to detect hidden defects in concrete structures in a contactless, autonomous, and efficient manner. The ResNet50 model was trained on simulated data and achieved a validation accuracy of 99.93% in classifying subsurface-defected and defect-free concrete blocks. Laboratory experiments involving the compression of concrete blocks and thermal imaging yielded a dataset used to train a new model with the same architecture and hyperparameters, resulting in a validation accuracy of 100%. This investigation demonstrates that AI can effectively classify thermal images of concrete surfaces, enabling the detection of subsurface cracks and hidden defects—an accuracy of 99.93% in classifying subsurface-defected and defect-free concrete blocks.

Laboratory experiments involving the compression of concrete blocks and thermal imaging yielded a dataset used to train a new model with the same architecture and hyperparameters, resulting in a validation accuracy of 100%. This investigation demonstrates that AI can effectively classify thermal images of concrete surfaces, enabling the detection of subsurface cracks and hidden defects.