• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• autumn
• /
• pp.180-185
• /
• 2003

The increase of traffic over a bridge has been emerged as one of the most severe problems in view of bridge maintenance, since the load effect caused by the vehicle passage over the bridge has brought out a long-term damage to bridge structure, and it is nearly impossible to maintain operational serviceability of bridge to user's satisfactory level without any concern on bridge maintenance at the phase of completion. Moreover, bridge maintenance operation should be performed by regular inspection over the bridge to prevent structural malfunction or unexpected accidents front breaking out by monitoring on cracks or deformations during service. Therefore, technical breakthrough related to this uninterested field of bridge maintenance leading the public to the turning point of recognition is desperately needed. This study has the aim of development on automated inspection system to lower surface of bridge superstructures to replace the conventional system of bridge inspection with the naked eye, where the monitoring staff is directly on board to refractive or other type of maintenance .vehicles, with which it is expected that we can solve the problems essentially where the results of inspection are varied to change with subjective manlier from monitoring staff, increase stabilities in safety during the inspection, and make contribution to construct data base by providing objective and quantitative data and materials through image processing method over data captured by cameras. By this system it is also expected that objective estimation over the right time of maintenance and reinforcement work will lead enormous decrease in maintenance cost.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.9-12
• /
• 2021

드론 Photogrammetry는 높은 기술적 활용가치가 있는 기술로서, 결과물로 생성하는 3D 디지털 공간정보 모델이 시설물의 비육안 안전점검 및 진단에 활용될 수 있을 뿐만 아니라 디지털 트윈 구축을 위한 가장 기초적이고 핵심적인 수치 데이터를 제공하기 때문이다. 본 연구에서는 드론 Photogrammetry의 적정 품질을 구현하기 위한 임무비행의 다양한 영향인자에 대해 고찰하였다. K-water연구원 누수탐사실습장을 대상으로 드론 사진 촬영 시 비행고도, 비행속도, 중첩도, 카메라 Pitch각의 영향에 대해 연구를 수행하였다. 본 연구에서 비행시간에 영향을 미치는 인자로서 비행고도, 중첩도, 비행속도의 순으로 중요도가 있음을 알 수 있었다. 드론 임무 비행 시 후처리 결과에 가장 큰 영향을 미치는 인자는 중첩도로 나타났다. 중첩도 60% 임무비행은 3D 모델의 geometry 왜곡이 큰 편으로 나타났다. 비행 고도는 GSD (Ground Sampling Distance)와 직접 연계되므로 중요하며, 낮은 고도일수록 높은 품질의 모델링이 가능하다. April Tag를 통한 지상기준점 자동 패턴 인식 기능은 후처리 과정에서 시간 절약이 가능하여 유용하였다. 비행속도에 의한 결과물의 품질은 큰 차이가 없었으나, 수직 구조물의 모서리 부분에 다소 차이가 있었다. 짐벌 Pitch각도에 의한 정사영상 품질의 차이는 크지 않았으나 수직구조물과 평면적 구조물에 따라 각기 다른 촬영각도를 적용하는 것이 바람직하다. 본 연구성과는 향후 보다 다양한 환경에서의 데이터 수집을 통해 최적 디지털 현실 모델링에 기여할 것으로 판단된다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.827-829
• /
• 2021

현대사회의 시설물 중 다수가 콘크리트를 사용하여 건설되었고, 재료적 성질로 인해 균열, 박락, 백태 등의 손상이 발생하고 있고 시설물 관리가 요구되고 있다. 하지만, 현재 시설물 관리는 사람의 육안 점검을 정기적으로 수행하고 있으나, 높은 시설물이나 맨눈으로 확인할 수 없는 시설물의 경우 관리가 어렵다. 이에 본 논문에서는 다양한 영상장비를 활용해 시설물의 이미지에서 균열을 분류하는 알고리즘을 제안한다. 균열 분류 알고리즘은 산업 이상 감지 데이터 세트인 MVTec AD 데이터 세트를 사전 학습하고 L2 auto-encoder를 사용하여 균열을 분류한다. MVTec AD 데이터 세트를 사전학습시킴으로써 균열, 박락, 백태 등의 특징을 학습시킬 수 있을 것으로 기대한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.45-48
• /
• 2022

건축물 안전 점검은 대부분 전문가의 현장 방문을 통한 육안검사다. 그중 균열 검사는 건물 위험도를 나타내는 중요한 지표로써 발생 위치, 진행성, 크기를 조사하는데, 최근 균열 조사 방식에 대해 객관성과 체계성을 보완할 딥러닝 개발이 활발하다. 그러나 균열 이미지는 외부 현장에 모양, 규모도 많은 종류라 도메인이 다양해야 하는데 대부분 제한된 환경과 실제적인 균열 검사와는 무관한 데이터로 구성되어 실효적이지 않다. 본 연구에서는 균열 조사에 적합하고 Wild 환경에 적용 가능한 POC 데이터셋을 소개한다. 기존 균열 공인 데이터셋 4종의 특징과 한계점을 분석을 토대로 고해상도 이미지로써 균열의 세부 특징을 담았고 균열 유사 환경과 조건들을 추가 촬영해 균열 검출에 강인하게 학습되도록 지향하였다. 정제 및 라벨링 작업을 거친 POC 데이터 셋은 균열 검출모델인 YOLO-v5으로 성능을 실험하였고, mAP(mean Average Precision) 75.5%로 높은 검출률을 보였다. POC 데이터셋으로 더욱 도메인에 적응적(Domain-adapted)인 인공지능 모델을 개발하여 건물, 댐, 교량 등 각종 대형 건축물에 대한 안전하고 효과적인 안전 관리 도구로써 활용할 것을 기대한다.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.9 no.3
• /
• pp.303-310
• /
• 2021

Cracks in bridges are important factors that indicate the condition of bridges and should be monitored periodically. However, a visual inspection conducted by a human expert has problems in cost, time, and reliability. Therefore, in recent years, researches to apply a deep learning model are started to be conducted. Deep learning requires sufficient data on the situations to be predicted, but bridge crack data is relatively difficult to obtain. In particular, it is difficult to collect a large amount of crack data in a specific situation because the shape of bridge cracks may vary depending on the bridge's design, location, and construction method. This study developed a crack detection model that generates and trains insufficient crack data through a Generative Adversarial Network. GAN successfully generated data statistically similar to the given crack data, and accordingly, crack detection was possible with about 3% higher accuracy when using the generated image than when the generated image was not used. This approach is expected to effectively improve the performance of the detection model as it is applied when crack detection on bridges is required, though there is not enough data, also when there is relatively little or much data f or one class.

• Journal of Korean Society of Disaster and Security
• /
• v.16 no.1
• /
• pp.23-35
• /
• 2023

Recently, the damage caused by natural disasters such as typhoons and localized torrential rains has been increasing rapidly. The Ministry of the Interior and Safety enacted a ｢law on safety management of small sized infrastructures｣ and local governments have to register small sized infrastructures with the National Disaster and Safety Management System (NDMS) until March 31st every year. Recently, each local government has ordered Safety inspections of small sized infrastructures and maintenance plans and six types of facilities, including small streams, small bridges, farm roads, access roads to village, inlet weirs, and drop structures are being surveyed and digitized into a database. Each facility is being evaluated for risk, and for those deemed hazardous, maintenance plans are being developed. However, since the risk assessment method of small sized infrastructures is not clear so that is conducted through visual investigation by field investigators, risk assessment is conducted in a subjective and ambiguous form. Therefore, this study presented a reasonable and quantitative risk assessment method by providing a quantitative evaluation indicator for small stream, which has the highest disaster risk among other small sized infrastructures, so that small sized hazard infrastructures can be selected to secure transparent evidence for improvement plans and action plans.

• Plant Journal
• /
• v.11 no.2
• /
• pp.39-44
• /
• 2015

According to a research that is based on a previous study, But in a different way, This study shows fault diagnosis of Induction motor bearing which runs in coal-fired power plant industries on Desulfurization absorber agitator using Spectrum analysis of Stator Current and visual inspection. As a result of harmonic content analysis of stator current spectrum, It was possible to detect ball and outer race fault frequency. The comparison in the context of this experiment proves that the amplitude of faulty frequency is increased in three times at a fault in ball and in outer race. Spectrum analysis of stator current can be used to detect the presence of a fault condition as well as experiment in faulty bearings, besides early fault detection in bearings can prevent unexpected power generation loss and emergency maintenance cost.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.10
• /
• pp.15-22
• /
• 2020

In this study, we propose an improvement method that can create U-Net model which detect fine concrete cracks by applying a weighted loss function. Because cracks in concrete are a factor that threatens safety, it is important to periodically check the condition and take prompt initial measures. However, currently, the visual inspection is mainly used in which the inspector directly inspects and evaluates with naked eyes. This has limitations not only in terms of accuracy, but also in terms of cost, time and safety. Accordingly, technologies using deep learning is being researched so that minute cracks generated in concrete structures can be detected quickly and accurately. As a result of attempting crack detection using U-Net in this study, it was confirmed that it could not detect minute cracks. Accordingly, as a result of verifying the performance of the model trained by applying the suggested weighted loss function, a highly reliable value (Accuracy) of 99% or higher and a harmonic average (F1_Score) of 89% to 92% was derived. The performance of the learning improvement plan was verified through the results of accurately and clearly detecting cracks.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.7
• /
• pp.1575-1581
• /
• 2005

Recently, further study is being done on the affect of crack on concrete structure and many people have made every endeavor not to leave it unsettled but to minimize it by repair works. In this paper we propose the image processing method that do not remain manual but automatically process the length, the direction and e width of cracks on concrete surface. First, we calibrate light's affect from image by using closing operation, one of morphology methods that can extract the feature of oracle and we extract the edge of crack image by sobel mask. After it, crack image is binarized by iteration binarization. And we extract the edge of cracks using noise elimination method that use an average of adjacent pixels by 3${\times}$3 mask and Glassfire Labeling algorithm. on, in this paper we propose an image processing method which can automatically measure the length, the direction and the width of cracks using the extracted edges of cracks. The results of experiment showed that the proposed method works better on the extraction of concrete cracks. Also our method showed the possibility that inspector's decision is unnecessary.

• Geophysics and Geophysical Exploration
• /
• v.16 no.4
• /
• pp.275-283
• /
• 2013

The sea dike is the most important facility of reclamation projects, and plays an important role in securing freshwater in the reservoir. Systematic research on practical approaches and data analysis techniques are lacking even though some geophysical methods such as electrical resistivity and self-potential surveys are included within the inspection processes. Hence, geophysical methods were considered for improvement of precision safety diagnosis methods after problems in the maintenance system have been identified, such as safety checks and precision safety diagnoses. In addition, geophysical methods customized according to variations in ambient environmental limiting factors such as pore pressure changes by tidal fluctuation, compaction characteristics of the fill materials, and the surface condition of the embankment were suggested.