The recent earthquake of Pohang (M5.4) and the Gyeongju earthquake (M5.8) suggested the possibility of a strong earthquake in Korea and reminded us that the Korea is no longer an earthquake-safe zone. In the disaster recovery stage in a disaster like an earthquake, the investigation of the damage situation and the safety assessment of the building serve to provide important information for the initial action such as establishment of the recovery strategy and rescue of the survivor. However, the research that depends on manpower can not cope with the difficulty of processing a large number of doses in a short time, and the expertise of the manpower must be taken into consideration, which may result in delayed initial action. The purpose of this study is to develop the SMART SKY EYE system for the rapid structural safety evaluation method of buildings using the unmanned aerial vehicle and the infrared module. After developing the necessary technology for the safety inspection using the unmanned aerial vehicle, I proposed SMART SKY EYE System which is an rapid structural safety evaluation technology for building using unmanned aerial vehicle and infrared module that evaluates the performance and safety of buildings by integrating existing safety inspection method and evaluation standard with unmanned aerial vehicle technology and constructed defect detection algorithm. I propose a 4-step check process for structural fault detection, displacement measurement, 3D reconstruction (3D representation), safety grade designation and evaluation. By using the unmanned aerial vehicle, the existing safety inspection process was reconstructed and the possibility of the inspection technology development was confirmed. Also in this paper, the proposed protocol was verified by applying it to existing old buildings, and defect information could be quantified by calculating length, width, and area for each defect. 3D modeling was performed using the aerial photographs obtained from the unmanned aerial vehicle and high resolution images of a level enough for visual inspection were obtained through image search. Compared with the visual inspection of the existing safety inspection, I confirmed the feasibility of defect tracking management through visual data collection and data accumulation.
Experimental study was carried out to develop concrete crack depth assessment method using infrared thermography. BIG DATA was constructed through data collection of thermal image and variables and analyzed through machine learning. Among the five algorithms constructed, the prediction accuracy of the crack depth was 72% when random forest (RandomForest) was implemented, and the prediction accuracy was 89% when it was recognized up to the adjacent depth (Class) Respectively. Therefore, this algorithm can confirm the possibility of crack depth information prediction by using thermal image. The crack depth information obtained by this algorithm is combined with the real image shown in the previous chapter and the damage information data (crack width, crack length, etc.) obtained through 3D modeling to construct the damage information integration system of the building including the crack depth information And can be used to build an emergency inspection platform using unmanned aerial vehicles.
In order to verify this, the buildings damaged by the earthquake in Pohang were checked and compared using this system. The factors necessary for the assessment of buildings damaged by earthquakes, such as appearance defects, slope and subsidence, and the risk of falling objects, were investigated and evaluated. As a result of the evaluation, all buildings showed a danger level, and the results are consistent with the results of the existing emergency earthquake risk assessment. As a result, the possibility of checking the emergency safety using the unmanned aerial vehicle for the damaged structures in case of a large-scale disaster such as an earthquake was confirmed. The results are consistent with the results of the existing emergency earthquake risk assessment. As a result, the possibility of checking the emergency safety using the unmanned aerial vehicle for the damaged structures in case of a large-scale disaster such as an earthquake was confirmed.