Infrared (IR) thermography has become a common technique for non-destructive inspections in various engineering fields. The IR thermography technique identifies and measures near surface defects by detecting the temperature gradient on the surface of a target object (e.g., a concrete wall). This study concerns the evaluation of the efficiency of the different types of NDT imaging methods of thermographic, colour, and greyscale imagery, in detecting various types of concrete damage. Data from the NDT imaging techniques are used to quantify near-surface structural damage by using image processing techniques, and validation of damage quantities is done with data from visual inspection. In terms of imagery, the thermographic classifications produced higher accuracies than the greyscale classifications. This is partly due to the fact that thermographic images contain less variability within the concrete imagery, and at, the same time increase the visibility of cracks that may be otherwise imperceptible. The results show good correlation between damage quantities obtained through the image processing techniques applied to the IR thermography and optical imaging and the results obtained through visual inspection using the Feeler gauge and crack detection microscope.
展开▼
