To identify the defects illustrated on building facades by employing infrared thermography under shadow

Abstract

Infrared thermography (IRT) has been employed to identify defects illustrated on building facades. However, environmental conditions (such as a shadow) do affect the IRT by making it difficult to identify defects. To estimate the shadow effects quantitively, this study has proposed an approach based on a multiplicated model. This then allows the IRT to be segmented into classes according to surface temperature information by employing an image segmentation technique. In doing so, the regional boundaries can be extracted. Furthermore, segmented regions with a higher surface temperature than that of their neighbors can be treated as defects. Acoustic tracing is another non-destructive method that has been employed. Acoustic tracing can be used to identify defects through acoustic response differences. When both sets of results are overlaid together, and the overlaid results are shown to highly coincide.