Crack detection in historical masonry structures using efficient image processing: Application on a masonry bridge in Iran

Historical masonry structures are frequently subjected to a state of damage or deterioration that leads growing concern for their conservation. Surveying the pre-existing state of crack is a preliminary paramount activity to follow the continuous evolution of damage, which is a central topic in structural health monitoring. The present contribution focuses on such preliminary step, proposing an innovative optic technique to identify with high accuracy the external crack patterns in existing masonry walls. The method is called derivation-based crack detection and works mainly on high definition pictures, to identify in post processing both major and minor cracks already present in a masonry façade. The technique is benchmarked on a historical heritage structure in Iran, namely the Khaju Bridge; from the analyses carried out, it is found that the method is simple and robust and that is able to identify accurately also less evident crack patterns with high accuracy.