Detection of damage in heritage constructions based on 3D point clouds. A systematic review

3D point clouds have emerged as a crucial source of information in the field of Cultural Heritage due to their ability to capture reality with high density and accuracy. They are commonly used as inputs for generating planimetries, Building Information Models or numerical simulations. However, their potential goes beyond these applications and can be further exploited, particularly in the realm of detecting and quantifying damage. Taking this into consideration, the aim of this study is to conduct a systematic review to compile previous experiences in which the 3D point cloud has been applied for the detection of damage in historical constructions. A total of 188 articles were analysed during this review, emphasizing the capability of this technology in detecting damage in historical constructions. The review revealed a growing trend in this approach, particularly in the structural field where the terrestrial laser scanner is the most commonly used sensor. This is due to the capacity of capturing reality with great density and accuracy, allowing for the extraction of different type of damages related to the geometry of the elements such as deflections or out-of-plane deformations. However, this is not the only valuable information. 3D point clouds, which can also be obtained through the Structure From Motion approach, are capable of storing radiometric information in the form of intensity and/or digital levels. This information can be utilized for detecting other types of damage, such as moisture or crusts, among others. All this information is properly synthetized and presented, in accordance with different aspects such as the general properties of the 3D point clouds, the use of geometric and radiometric features, the integration of additional sensors, and their use in other approaches. Additionally, the work proposes a novel classification system for damage detection based on 3D point clouds. This system is based on the type of feature/s used.