RESEARCH
BUILDING FORENSICS
Machine Learning-Based Concrete Crack Depth Prediction Using Thermal Images Taken under Daylight Conditions
The non-destructive and non-contact testing method was introduced using thermal images and machine learning to detect concrete crack depth. The thermal images of the cracked specimens were obtained using a constant test setup for several months under daylight conditions, which provided sufficient heat for measuring the temperature distributions of the specimens, with recording parameters such as air temperature, humidity, and illuminance. The thermal images obtained the crack and surface temperatures depending on the crack widths and depths using the parameters. The results of crack depth prediction were compared to identify the best algorithm. In order to confirm machine learning efficiency, data bias analysis was conducted using principal component analysis, singular value decomposition, and independent component analysis.
SMART SKY EYE System for Preliminary Structural Safety Assessment of Buildings Using Unmanned Aerial Vehicles
An aerial image-based approach was used to inspect cracks and deformations in buildings. A state-of-the-art safety evaluation method termed SMART SKY EYE (Smart building safety assessment system using UAV) is introduced; this system utilizes an unmanned airplane equipped with a thermal camera and programmed with various surveying efficiency improvement methods, such as thermography, machine-learning algorithms, and 3D point cloud modeling. This method can obtain crack maps, crack depths, and the deformations of structures. Error rates are compared between the proposed and conventional methods.
Research and development of application technologies that combine high-tech technologies have been actively conducted in the Fourth Industrial Revolution. Building information modeling (BIM) technology using advanced equipment is promising for future construction projects. In particular, using a 3D laser scanner, LiDAR is expected to be a solution for future building safety inspections. This work proposes a new method for evaluating building stability using a 3D laser scanner. This study analyzed an underground parking lot using a 3D laser scanner. Further, structural analysis was performed using the finite element method (FEM) by applying the figure and geometry data acquired from the laser scan. This process includes surveying the modeled point cloud data of the scanned building, such as identifying the relative deflection of the floor slab and the sectional shape and inclination of the column. Consequently, safety diagnosis was performed using the original evaluation criteria. This study demonstrates the precision and efficiency of using a 3D laser scanner for assessing building stability, presenting a digital point cloud-based approach.
