• 인터넷정보학회논문지
• /
• 제18권5호
• /
• pp.69-77
• /
• 2017

본 논문에서는 재난상황에서 딥러닝과 증강현실 기술을 활용한 재난대응 방안과 그에 따른 네트워크 자원 확보 방안을 제안한다. 딥러닝과 증강현실 기술의 특징과 현황을 파악하고, 재난분야와의 연관성에 관하여 설명한다. 딥러닝 기술을 사용하여 재난 상황을 정확하게 인지하고 관련 재난 정보를 증강현실로 구현하여 재난대응 현장 및 통합지원본부, 재난안전대책본부 등에 제공함으로써 재난대응 역량을 강화할 수 있다. 각종 재난사례 중 화재상황을 중점으로, 딥러닝 기반 화재상황 인식 및 증강현실 정보제공을 통해 효과적으로 재난대응 역량을 강화할 수 있는 방안을 제시한다. 마지막으로, 본 논문의 재난대응 방안을 활용하기 위한 네트워크 자원 확보 기법을 제시한다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2022년도 학술발표회
• /
• pp.145-145
• /
• 2022

Availability of abundant water resources data in developing countries is a great concern that has hindered the adoption of deep learning techniques (DL) for disaster prevention and mitigation. On the contrary, over the last two decades, a sizeable amount of DL publication in disaster management emanated from developed countries with efficient data management systems. To understand the current state of DL adoption for solving water-related disaster management in developing countries, an extensive bibliometric review coupled with a theory-based analysis of related research documents is conducted from 2003 - 2022 using Web of Science, Scopus, VOSviewer software and PRISMA model. Results show that four major disasters - pluvial / fluvial flooding, land subsidence, drought and snow avalanche are the most prevalent. Also, recurrent flash floods and landslides caused by irregular rainfall pattern, abundant freshwater and mountainous terrains made India the only developing country with an impressive DL adoption rate of 50% publication count, thereby setting the pace for other developing countries. Further analysis indicates that economically-disadvantaged countries will experience a delay in DL implementation based on their Human Development Index (HDI) because DL implementation is capital-intensive. COVID-19 among other factors is identified as a driver of DL. Although, the Long Short Term Model (LSTM) model is the most frequently used, but optimal model performance is not limited to a certain model. Each DL model performs based on defined modelling objectives. Furthermore, effect of input data size shows no clear relationship with model performance while final model deployment in solving disaster problems in real-life scenarios is lacking. Therefore, data augmentation and transfer learning are recommended to solve data management problems. Intensive research, training, innovation, deployment using cheap web-based servers, APIs and nature-based solutions are encouraged to enhance disaster preparedness.

• 국제학술발표논문집
• /
• The 4th International Conference on Construction Engineering and Project Management Organized by the University of New South Wales
• /
• pp.582-585
• /
• 2011

During disaster response, prioritization of limited resources is one of the most important bust challenging tasks. At the same time, it is imperative to timely provide the rescuers with the adequate equipment to facilitate lifesaving operations. However, supply of high demand equipment was insufficient during the initial phase of disaster response, challenging lifesaving operations in the case of the 9-11 terrorist attacks. In respond to the Haiti Earthquake, spatial information of the geographic area was not sufficient to support the search and rescue operations in the early phase of disaster response. However, with the help of civilians, information such as road names, infrastructure damage, and victim locations were updated into the spatial data repository. At the same time, resource outside of the disaster affected zone converges into the area to assist the response efforts, which is the effect of convergence that often made resource coordination challenging in large scale disasters. To efficiently collect information and utilize the converging resources, this paper proposes a flexible data repository for information update for equipment utilization in large scale disaster response scenarios.

• 한국재난정보학회 논문집
• /
• 제7권2호
• /
• pp.140-150
• /
• 2011

지구 온난화로 생활 환경이 급격히 변화하고 있으며, 대형 재난이 증가하고 있다. 이러한 재난 발생시 복구에 많은 자원을 투입하고 있지만, 재난의 예방 만큼 효과적인 대책은 없을 것이다. 재난전조 정보란 하인리히 법칙에 따라 예고되는 재난에 대한 전조이며, 이에 대한 정보를 자동으로 추출하여 대비할 수 있게 하는 것이 본 논문의 초점이다. 웹에 산재된 정보로부터 전조 정보를 정확히 추출하기 위한 기반이 되는 단어(명사)를 구축하고 이를 기반으로 정확한 데이터를 추출할 수 있는 알고리즘을 연구하였다. 본 연구의 결과물로 도출된 단어는 분석적인 연구결과이기 때문에 장기적으로 실제 데이터를 적용하면서 지속적으로 보완되어야 할 것이다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2004년도 학술발표회
• /
• pp.111-123
• /
• 2004

Recently, though damage caused by intensive rainfall and typhoon happens frequently, we could not forecast or predict a disaster, due to the difficulty of obtaining exact information about it. For efficient disaster management, the most urgent need is the preparation of a flood forecast-warning system. Therefore, we need to provide a program that has the ability of inundation analysis and flood forecast-warning using a geographic information system, and using domestic technology rather than that from foreign countries. In this research, we constructed a FDMS(Flood Disaster Management System) that is able to analyze real-time inundation data, and usins the GIS(Ceographic Information System) with prompt analyzing of hydrologic-topographical parameters and runoff-computation. Moreover, by expressing inundation analysis in three-dimensions, we were able to get to the inundation area with ease. Finally, we expect that the application of this method in the (food forecast-warning system will have great role in reducing casualties and damage.

• Water Engineering Research
• /
• 제4권4호
• /
• pp.187-192
• /
• 2003

The rainfall observation systems have largely been improved in Japan. The Japan Meteorological Agency, prefecture governments, and other administrative bodies have also increased the number of rain gauges thru out the country. The density of observatories is now one per several $\km^2$. Heavy rainfall information systems have been improved. Besides it, the Internet was popularized in the late 1990s, and has been used to transmit data of heavy rainfall. Internet accessible cellular phones have been popular in Japan since 1999. Such phones are expected to be useful in the field of disaster warning announcements, because they can automatically notify users bye-mail of pending disasters. The use of the Internet during natural disasters is groundbreaking in Japan today. However, in order to use disaster information effectively on Internet it is necessary to investigate how to use the information during the rainfall disaster. Therefore in our study we suggest methods on the effective construction and their use of information technology on Internet.

• 한국전자거래학회지
• /
• 제20권4호
• /
• pp.227-239
• /
• 2015

자연재해로 인한 사회기반시설물 피해를 저감시키기 위해서는 피해발생 초기 대응 및 신속한 복구가 중요하나, 현재는 현장인력의 상황전파에 초점을 맞추고 있으며 긴급 복구를 수행하기 위한 기술적 지원은 부족한 실정이다. 본 논문에서는 피해상황에 적합한 복구 방법을 도출하여 효율적인 자원분배를 수행하기 위한 웹기반 재난피해복구지원시스템 및 신속한 복구 자원을 요청하기 위한 전자문서 개발에 대해 설명한다. 획득된 현장의 피해정보는 미리 구축된 피해복구 시나리오에 따라 적절한 피해복구 공종 및 공법을 도출하게 되며, 복구 수행을 위한 자원을 할당하여 전자문서를 거쳐 신속하게 해당 복구자원 지원을 요청하도록 하였다. 필요한 복구자원의 식별과 신속한 전파는 재해의 대응과 긴급복구에 큰 도움이 될 것으로 판단된다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2005년도 학술발표회(2)
• /
• pp.1100-1101
• /
• 2005

• 한국컴퓨터정보학회논문지
• /
• 제21권2호
• /
• pp.43-49
• /
• 2016

In this paper, we propose the controllability function and service design to efficiently implement the control of the disaster communication network, using Internet of Things(IoT) Technology. In accordance with the spread of Information Communication Technology(ICT), the era for building a disaster communications system without exclusion over the entire areas has emerged. National wireless mesh networks for public safety and disaster relief have been evolving to strengthen the pre-disaster response system using the latest technologies through the convergence of various technologies and services from the viewpoint of the command and control between disaster response agencies. In line with such a technological paradigm shift, the controllability of the objects in the IoT has been emerging as a key quality requirement of a disaster communications system. In this study, the objects are classified by the subject of control according to the IoT component, such as data, network resources and services in order to effectively implement their controllability. In addition, based on the destination of this controllability, technologies and services have been designed that can reduce the damage caused by disasters. Technologies and services that were derived from this study must be implemented in the current disaster safety network systems together with the establishment of an infrastructure for the networks in order that all persons are able to effectively utilize the disaster communications system for their safety.

• 한국수자원학회논문집
• /
• 제52권12호
• /
• pp.963-973
• /
• 2019

본 연구는 도시침수 위험기준이 산정되지 않은 지역의 예·경보 기준을 예측하기 위해 유역특성 자료와 피해이력 기반으로 산정된 한계강우량을 활용하여 도시침수 위험기준을 추정하는 모델을 검토하였다. 위험기준 추정모델은 머신러닝 알고리즘의 하나인 Support Vector Machine을 이용하여 설계하였으며, 학습자료는 지역별 한계강우량과 유역특성으로 구성하였다. 학습자료는 정규화 한 후 SVM 알고리즘에 적용하였으며, SVM에 적용시 Leave-One-Out과 K-fold 교차검증 알고리즘을 이용하여 절대평균오차와 표준편차를 계산한 후 모델의 성능을 평가하였다. Leave-One-Out의 경우 표준편차가 작은 모델이 최적모델로 선정되었으며, K-fold의 경우 fold의 개수가 적은 모델이 선정되었다. 선정된 모델의 지속시간별 평균 정확도는 80% 이상으로 나타나 침수 위험기준 추정을 위해 SVM을 활용가능 할 것으로 판단된다.