• 지구물리
• /
• 제8권1호
• /
• pp.7-14
• /
• 2005

Early detection in real-time response of slope movements ensures tremendous saving of lives and repair costs from catastrophic disaster. Therefore, it is essential to constantly monitor the performance and integrity of slope-stabilizing structures such as Rock bolt, Nail and Pile during or after installation. We developed a novel monitoring system using Fiber Bragg Grating (FBG) sensor. It's advantages are highly sensitivity, small dimension and electro-magnetic immunity. capability of multiplexing, system integrity, remote sensing - these serve real-time health monitoring of the structures. Real-time strain measurement by the signal processing program is shown graphically and it gives a warning sound when the monitored strain state exceeds a given threshold level so that any sign of abnormal disturbance on the spot can be easily perceived.

• 대한토목학회논문집
• /
• 제38권3호
• /
• pp.439-448
• /
• 2018

지진 조기경보 시스템(EEWS)은 상대적으로 빠른 속도로 전파되는 P파를 관측하여 잇따라 도달하는 S파에 의한 피해 위험을 미리 경보하고, 시민의 즉각적인 대피를 유도하는 것을 목표로 한다. 현재 대한민국의 지진 조기경보 시스템은 최초 P파 관측 후 50초 이내에 지진 경보 발령이 가능한 상태이며 2020년까지 지진 경보 발령 시간을 P파 관측 후 10초 이내로 단축하겠다고 공표한 상태이다. 이를 위해서는 지진 발생 시 P파의 관측이 5초 내에 이루어져야 한다. 2016년 리히터 규모 5.8의 경주 지진 이후 국내 지진 재난 대응체계에 대한 개선의 요구에 힘입어 재난 및 안전관리 기본법 내 재난관리책임기관의 기능 연속성 계획 수립에 관한 항이 신설되었다. 국립공원 관리공단 역시 재난으로부터의 탐방객 안전확보의 의무를 지닌 바, 보다 능동적인 대응체계를 위해 유관기관과의 협력체계 구축을 추진하고 있다. 이에 따라, 본 연구에서는 통합지진관측망의 분포 및 국립공원의 공간적 분포를 정량화하여 국립공원의 지진 조기경보 서비스 제공의 잠재적인 취약성을 분석하였다. 분석 결과, 중부지방에 위치하는 속리산, 계룡산, 가야산, 덕유산 및 동남부 지방의 경주 국립공원은 자체적으로 지진파를 감지하여 분석하는 현장시스템의 보강이 필요할 것으로 나타났으며 북한산 국립공원은 통합지진관측망 중심의 전방탐지시스템의 개선이 필요할 것으로 나타났다.

• 한국컴퓨터정보학회논문지
• /
• 제28권12호
• /
• pp.67-77
• /
• 2023

본 연구는 스마트팜 환경에서 진행된 혁신적인 연구로, 딥러닝을 기반으로 한 질병 및 해충 탐지 모델을 개발하고, 이를 지능형 사물인터넷(IoT) 플랫폼에 적용하여 디지털 농업 환경 구현의 새로운 가능성을 탐색하였다. 연구의 핵심은 Pseudo-Labeling, RegNet, EfficientNet 등 최신 ImageNet 모델과 전처리 방식을 통합하여, 복잡한 농업 환경에서 다양한 질병과 해충을 높은 정확도로 탐지하는 것이었다. 이를 위해 앙상블 학습 기법을 적용하여 모델의 정확도와 안정성을 극대화했으며, 평균 정밀도(mAP), 정밀도, 재현율, 정확도, 박스 손실 등의 다양한 성능 지표를 통해 모델을 평가하였다. 또한, SHAP 프레임워크를 활용하여 모델의 예측 기준에 대한 깊은 이해를 도모하였고, 이를 통해 모델의 결정 과정을 보다 투명하게 만들었다. 이러한 분석은 모델이 어떻게 다양한 변수들을 고려하여 질병 및 해충을 탐지하는지에 대한 중요한 통찰력을 제공하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권12호
• /
• pp.3904-3922
• /
• 2022

As a research hotspot, pedestrian detection has a wide range of applications in the field of computer vision in recent years. However, current pedestrian detection methods have problems such as insufficient detection accuracy and large models that are not suitable for large-scale deployment. In view of these problems mentioned above, a lightweight pedestrian detection and early warning method using a new model called you only look once (Yolov5) is proposed in this paper, which utilizing advantages of Yolov5s model to achieve accurate and fast pedestrian recognition. In addition, this paper also optimizes the loss function of the batch normalization (BN) layer. After sparsification, pruning and fine-tuning, got a lot of optimization, the size of the model on the edge of the computing power is lower equipment can be deployed. Finally, from the experimental data presented in this paper, under the training of the road pedestrian dataset that we collected and processed independently, the Yolov5s model has certain advantages in terms of precision and other indicators compared with traditional single shot multiBox detector (SSD) model and fast region-convolutional neural network (Fast R-CNN) model. After pruning and lightweight, the size of training model is greatly reduced without a significant reduction in accuracy, and the final precision reaches 87%, while the model size is reduced to 7,723 KB.

• 전기학회논문지P
• /
• 제66권2호
• /
• pp.82-87
• /
• 2017

This paper presents performance improvement of CO gas sensor signal conditioner for early fire warning system. The warning system is based on the CO sensor and its advanced signal conditioning modules network that employ electochemical gas sensor. The electochemical has advantage of having a linear output and operating with a low consumption and fast response. This electrochemical gas sensor contains a gas membrane and three electrodes(working, counter, reference electrode) in contact with an electrolyte. To use a three-electrode sensor, a voltage has to be applied between the working and the reference electrode according to the specification of the sensor. In this paper, we designed these requirements that should be considered in temperature compensation algorithm and electrode measurement of CO sensor modules by using advanced signal conditioning method included 3-electrode. Simulation and experimental results show that signal conditioner of CO sensor module using 3-electrode have a advantage linearity, sensitivity and stability, fast response etc..

• 경영정보학연구
• /
• 제12권2호
• /
• pp.89-104
• /
• 2010

최근 유비쿼터스 센서 네트워크(Ubiquitous Sensor Network: USN)는 환경 모니터링을 비롯한 다양한 분야에 적용되어 왔다. 몇몇 연구는 재난 및 방재 분야, 특히 문화재관리 분야에 USN을 적용하기도 하였다. USN은 문화재의 훼손과 소실의 원인이 되는 화재를 조기탐지하기 위해 실시간 온라인으로 모니터링하는 유용한 기술이다. 특히, 인간이 접근하기 어렵거나, 외관이나 미관을 중시하는 문화재의 경우는 USN을 적용하여 이를 모니터링하는 것이 필수적이다. 그러나 인간이 화재 발생 현장을 컴퓨터 화면으로 직접 관찰하지 않고 USN으로부터 수집된 데이터만으로 화재 발생 여부를 자동으로 판별하는 데는 늘 오보의 한계점이 존재한다. 본 연구에서는 이러한 한계점을 해결하기 위해 온톨로지를 적용하는 방안을 제시하였고, 실험실 환경에서 USN으로부터 수집된 데이터를 기초로 화재 발생 여부를 조기에 더욱 정확하게 탐지할 수 있는 온톨로지와 추론규칙을 설계하여 실험하였다.

• 한국해양공학회지
• /
• 제33권6호
• /
• pp.607-614
• /
• 2019

It has been an issue among researchers that the tsunamis that occurred on the west coast of Japan in 1983 and 1993 damaged the coastal cities on the east coast of Korea. In order to predict and reduce the damage to the Korean Peninsula effectively, it is necessary to install offshore tsunami observation instruments as part of the system for the early detection of tsunamis. The purpose of this study is to recommend the optimal deployment of tsunami observation instruments in terms of the higher probability of tsunami detection with the minimum equipment and the maximum evacuation and warning time according to the current situation in Korea. In order to propose the optimal location of the tsunami observation equipment, this study will analyze the tsunami propagation phenomena on the east sea by considering the potential tsunami scenario on the west coast of Japan through numerical modeling using the COrnell Multi-grid COupled Tsunami (COMCOT) model. Based on the results of the numerical model, this study suggested the optimal deployment of Korea's offshore tsunami observation instruments on the northeast side of Ulleung Island.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2006년도 Proceedings of ISRS 2006 PORSEC Volume II
• /
• pp.827-830
• /
• 2006

The tsunami from the megathrust earthquake magnitude 9.3 on 26 December 2004 is the largest tsunami the world has known in over forty years. This tsunami destructively attacked 13 countries around Indian Ocean with at least 230,000 fatalities, displaced people 2,089,883 and 1.5 million people who lost their livelihoods. The ratio of women and children killed to men is 3 to 1. The total damage costs US$ 10.73 billion and rebuilding costs US$ 10.375 billion. The tsunami's death toll could have been drastically reduced, if the warning was disseminated quickly and effectively to the coastal dwellers along the Indian Ocean rim. With a warning system in Indian Ocean similar to that operating in the Pacific Ocean since 1965, it would have been possible to warn, evacuate and save countless lives. The best tribute we can pay to all who perished or suffered in this disaster is to heed its powerful lessons. UNESCO/IOC have put their tremendous effort on better disaster preparedness, functional early warning systems and realistic arrangements to cope with tsunami disaster. They organized ICG/IOTWS (Indian Ocean Tsunami Warning System) and the third of this meeting is held in Bali, Indonesia during $31^{st}$ July to $4^{th}$ August 2006. A US$ 53 million interim warning system using tidal gauges and undersea sensors is nearing completion in the Indian Ocean with the assistance from IOC. The tsunami warning depends strictly on an early detection of a tsunami (wave) perturbation in the ocean itself. It does not and cannot depend on seismological information alone. In the case of 26 December 2004 tsunami when the NOAA/PMEL DART (Deep-ocean Assessment and Reporting of Tsunami) system has not been deployed, the initialized input of sea surface perturbation for the MOST (Method Of Splitting Tsunami) model was from the tsunamigenic-earthquake source model. It is the first time that the satellite altimeters can detect the signal of tsunami wave in the Bay of Bengal and was used to validate the output from the MOST model in the deep ocean. In the case of Thailand, the inundation part of the MOST model was run from Sumatra 2004 for inundation mapping purposes. The medium and high resolution satellite data were used to assess the degree of the damage from Indian Ocean tsunami of 2004 with NDVI classification at 6 provinces on the Andaman seacoast of Thailand. With the tide-gauge station data, run-up surveys, bathymetry and coastal topography data and land-use classification from satellite imageries, we can use these information for coastal zone management on evacuation plan and construction code.

• 해양환경안전학회지
• /
• 제23권6호
• /
• pp.684-689
• /
• 2017

본 연구는 참굴(Crassostrea gigas)의 패각운동을 이용하여, 연안역에서 발생하는 저염수에 대한 조기경보가능성을 살펴보았다. 30 psu와 20 psu에서 패각운동은 각각 $7.32{\pm}3.21$회/hr와 $7.11{\pm}3.90$회/hr였으며, 파형과 횟수는 차이가 없었다(t-test, p>0.001). 하지만 10 psu와 5 psu에서는 모든 개체가 폐각상태를 지속하였다. 수온과 염분의 복합실험결과, Group 1(수온 $15^{\circ}C$ ${\times}$ 염분 15 psu)은 20~30 psu에서 보인 패각운동 후(약 2~3시간), 장시간 폐각을 하였다. Group 2(수온 $30^{\circ}C$ ${\times}$ 염분 15 psu)에서는 Group 1의 패각 개폐운동보다 더 빠르고 자주 나타나, 참굴의 생리적인 위기상황에 대한 신호를 나타내었다. 따라서 이러한 파형은 하계 저염수 출현 시 나타낼 수 있는 조기경보 신호로 충분히 활용될 수 있을 것으로 보인다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권10호
• /
• pp.5006-5022
• /
• 2017

Effect and robust detection of targets in infrared images has crucial meaning for many applications, such as infrared guidance, early warning, and video surveillance. However, it is not an easy task due to the special characteristics of the infrared images, in which the background clutters are severe and the targets are weak. The recent literature demonstrates that sparse representation can help handle the detection problem, however, the detection performance should be improved. To this end, in this text, a hybrid method based on local sparse representation and contrast is proposed, which can effectively and robustly detect the infrared targets. First, a residual image is calculated based on local sparse representation for the original image, in which the target can be effectively highlighted. Then, a local contrast based method is adopted to compute the target prediction image, in which the background clutters can be highly suppressed. Subsequently, the residual image and the target prediction image are combined together adaptively so as to accurately and robustly locate the targets. Based on a set of comprehensive experiments, our algorithm has demonstrated better performance than other existing alternatives.