The likelihood of failure for the stress corrosion cracking (SCC) of caustic cracking, which affect to a risk of facilities, was analyzed through the risk based-inspection using API-581 BRD. We found that SCC of the caustic cracking was occurred above 5 % NaOH concentration, and the technical module subfactor (TMSF) was maximized for above 50 % concentration. The heat traced and monitoring were not sensitive to the TMSF with NaOH concentration and temperature. But the steam out was more of less affect minimum value of the TMSF. Also, the inspection number, the inspection effectiveness, and the year since inspection were very sensitive to the TMSF with NaOH concentration and temperature. Therefore, the plan of next inspection will be established with compositively considering those at once.

Bridge inspection based unmanned aerial vehicle (UAV) has received considerable attention due to its several advantages such as reliability and safety as well as saving time and cost. An unmanned inspection equipment for bridge inspection is composed of UAV and imaging devices such as RGB cameras and infrared cameras. However, many challenging issues should be solved in order to apply this technology to the field. In this paper, an UAV based crack detection method is investigated. To detect the cracks, the image processing techniques with deep learning algorithm are used. To build the spatial information of aging bridge, 3D point cloud based background model is generated.

These days, inexpensive and high efficiency UAV(Unmanned Aerial Vehicles) of disaster prevention and spatial information has been given more attention. But studies about test of accuracy of UAV were not enough despite high interest. UAV technology has been developed for military purposes, but in recent years, it has been used for various purposes in different forms in the private sector as spread to the public sector. However, since the study applying the UAV technologies to the inspection and maintenance of structures are insufficient so far, this study suggests an application method of UAV technology for inspection and investigation of infrastructure facilities. If the UAV technology is used in inspection and investigation of infrastructure facilities, it will be able to conduct a safety check and inspect high areas and narrow passages, which are difficult or impossible to access by manpower, and further improve the safety of the structure.

본 연구는 균열을 인식할 수 있는 영상기반 균열검사 알고리즘을 개발하고 성능을 검증하기 위한 모의 균열을 생성하는 방법에 관한 것이다. 균열검사 알고리즘을 개발하기 위해서는 균열 영상 데이터가 필요하다. 이러한 균열 영상들은 실제 균열이 발생한 콘크리트 벽면을 촬영하거나 일정한 하중을 가해 시편에 균열을 발생시키는 방법으로 획득하고 있다. 하지만 균열을 검사하기 위한 영상처리 알고리즘을 검증하기 위해서는 균열 폭이 일정한 정량화된 균열 영상이 필요하다. 이러한 정량화된 균열을 생성하기 위하여 크랙게이지와 같이 검은 명암을 가지는 표면 부착형 모의 균열이 사용되기도 한다. 표면 부착형 모의 균열을 사용하면 원하는 형태와 크기를 가지는 균열을 묘사하기 유리하며, 다양한 콘크리트 표면에 부착하여 균열 영상을 획득하고 이를 균열검사 프로그램의 개발에 활용할 수 있다. 특히 균열의 정확한 크기 측정이나 인식률을 정량적으로 검증하는데 매우 효과적이다. 하지만 표면 부착형 모의 균열과 실제 균열의 유사성에 대해서는 아직 검증되지 않았다. 본 연구에서는 표면 부착형 모의 균열과 콘크리트 시편에 유도한 균열을 제작하여, 영상 촬영한 후 명암, 질감, 콘트라스트, 선명도를 분석하여 두 균열간의 특성을 분석하고 유사성을 검증하였다. 영상은 다양한 거리와 조도에서 촬영하여 거리와 조도의 변화와 두 가지 균열의 특성 변화와의 관계를 알아보았다.