• 한국항행학회논문지
• /
• 제13권6호
• /
• pp.1011-1016
• /
• 2009

본 논문에서는 공공장소에 광범위하게 설치되어있는 CCTV의 감시 기능을 활용하여 화재 발생 감지 방법을 제안하였다. 제안한 방법은 고정된 카메라로부터 칼라 정보를 이용하여 비디오 시퀀스의 화재 프레임 후보를 찾아내고, 공간 기법을 기반으로 감지된 화재 정보의 전경 색상을 분석하였다. 실험 결과, 비디오 시컨스에서 시 공간적 화재 후보 정보들이 급격히 변화할 때, 화재 감지의 성능이 우수함을 확인할 수 있었다.

• 한국화재소방학회논문지
• /
• 제32권6호
• /
• pp.40-45
• /
• 2018

전방위 감시카메라에는 어안렌즈를 이용하여 광대역감시가 가능하도록 물체감지알고리즘으로 물체를 단위별 레벨링한 다음 전방위 감시카메라와 추적(PTZ)카메라로 구성된 시스템으로 연동하여 현장실험한 것이다. 전방위 감시카메라가 움직이는 물체를 정확히 감지하며 사각표시를 하였고 추적카메라와 유기적으로 연동하며 확대 추적하였다. 감지카메라와 화염감지 및 온도에 대한 현장실험에서는 오토스캔 중 화염이 감지되면 멈추며, 해당 화점부분을 화면의 중심부분으로 이동시켜 온도가 표출되었다. 또한 화염이격거리별 검지에 필요한 발열량의 인정기준인 1 km 2,340 kcal를 초과한 1.5 km에서도 가능하였다. 거리에 따른 화염감지성능시험에서는 거리 1 km일 때 폭 56 cm ${\times}$ 높이 90 cm를 초과한 1.5 km에서도 가능하여 산불화재에도 적응성이 충분하였다. 향후 석유 가스비축시설 및 저유소에 설치하면 자체는 물론 주위 화재예방 및 침입감시 등의 안전에 매우 유용할 것으로 기대된다.

• 한국안전학회지
• /
• 제36권2호
• /
• pp.87-93
• /
• 2021

Modern society shows rapid growth that is different from that of the development of existing technologies. The development of these technologies has led to the tendency of buildings to become dense, large and advancing. Regarding fire hazards, the possibility of large-scale fires causing fatal damage, due to the rapid spread of fire, increases. Therefore, for this reason, fire defense, i.e. detection and fire extinguishing facilities, in buildings are essential and well applied. But there are always limitations to that. Based on this reason, we would like to suggest the introduction of a new concept of a fire safety system. The method presented here is not only to use a single system for fire detection and fire extinguishing systems but to jointly use it in the environment and energy management fields within the building. However, an important step is required before introducing a system of these technologies. The fire extinguishing method proposed by this system is a method of extinguishing by blocking oxygen flowing into the space where the fire occurred. However, a sufficient basis is needed for this system to be applied in practice. Therefore, in this study, we intend to conduct a preliminary experiment to introduce the new concept of fire detection and extinguishing. The experiment used ethanol with a relatively simple combustion reaction and a high possibility of complete combustion. As a result, it was confirmed how the internal values changed during a fire using ethanol. Resultingly, we obtained the internal oxygen concentration and internal environmental changes according to the initial flame size. Lastly, the data accumulated in this study can be used as data for application in an automatic fire extinguishing system.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제5권11호
• /
• pp.607-614
• /
• 2016

본 논문에서는 컨볼루션 신경망 모델을 이용한 지능형 화재 학습 및 탐지 시스템을 제안한다. 제안하는 시스템에서 사용된 신경망의 컨볼루션 층을 통해 불꽃 이미지와 연기 이미지에 대한 특징맵을 생성하고, 생성된 특징맵에 대하여 불꽃과 연기를 분류하는 학습을 진행한다. 이렇게 학습된 신경망에 움직임 특징 및 색상 특징만을 이용한 간단한 처리를 통해 검출된 화재 후보 영역 이미지를 입력시키면 입력된 영역에 화재가 발생했는지의 여부를 알 수 있다. 다양한 영상을 대상으로 실험한 결과 학습된 신경망은 화재 후보 영역에서 불꽃과 연기를 분류하는데 뛰어난 효과를 보여줌을 확인하였다.

• 한국자동차공학회논문집
• /
• 제8권2호
• /
• pp.19-26
• /
• 2000

This paper presents the effects of initial pressure of mixture on CO, $CO_2$ and NOx emissions in constant volume combustion chamber. The CO, $CO_2,O_2,N_2$ concentrations in the chamber are determined by thermal conductivity detection (Gas-chromatograph) wile the NOx concentration is measured by chemiluminescent detection (NOx Analyser). Methane-air mixture is used as premixed fuel and the measurements are taken with equivalence ratios($\phi$) varing from 0.6 to 1.3, and initial pressures of methane-air mixture varing from 0.1MPa to 0.8MPa in constant volume combustion chamber. The NOx concentration steadily increases with increasing equivalence ratio, peaks in lean flame ($\phi$=0.85~0.9), and then rapidly decreases. However, as the initial pressure of mixture is increased, the equivalence ratio corresponding to the point of peak [NOx] shifts towards leaner conditions. This is caused by a similar shift in the peak [CH], which is caused by the variation with pressure and equivalence ratio of the rate of CH production from $CH_2$ and OH. The maximum combustion pressure peaks at $\phi$ =1.05 and the $CO_2$ concentration peaks at $\phi$=0.95~1.0 while the CO concentration rises sharply at the condition of fuel-rich mixtures. This is caused by complete combustion at $\phi$=0.95.

• 한국멀티미디어학회:학술대회논문집
• /
• 한국멀티미디어학회 2012년도 춘계학술발표대회논문집
• /
• pp.284-286
• /
• 2012

There has been increasing attention about the prevention and counter-measure of disasters. Particularly, for the case of fire disaster, early detection reduces the damage caused by fire significantly and effective detection method is important. Since most existing detectors need to be located at a close distance to fire, analyzing camera images to find fire becomes active research topic. In this paper, we analyze the color characteristics of fire images in various color spaces and report the experimental detection results. The best result is 77.8% success rate in YIQ space.

• 한국군사과학기술학회지
• /
• 제26권3호
• /
• pp.234-245
• /
• 2023

The Arena Fragmentation Test(AFT) is designed to analyze warhead performance by measuring fragmentation data. In order to evaluate the results of the AFT, a set of AFT images are captured by high-speed cameras. To detect objects in the AFT image set, ResNet-50 based Faster R-CNN is used as a detection model. However, because of the low resolution of the AFT image set, a detection model has shown low performance. To enhance the performance of the detection model, Super-resolution(SR) methods are used to increase the AFT image set resolution. To this end, The Bicubic method and three SR models: ZSSR, EDSR, and SwinIR are used. The use of SR images results in an increase in the performance of the detection model. While the increase in the number of pixels representing a fragment flame in the AFT images improves the Recall performance of the detection model, the number of pixels representing noise also increases, leading to a slight decreases in Precision performance. Consequently, the F1 score is increased by up to 9 %, demonstrating the effectiveness of SR in enhancing the performance of the detection model.

• Bulletin of the Korean Chemical Society
• /
• 제27권6호
• /
• pp.875-880
• /
• 2006

A new, simple and fast flow injection analysis (FIA) method has been developed for the indirect determination of nitrite. The proposed indirect automatic method is based on the oxidation of nitrite to nitrate using a lead(IV) dioxide oxidant microcolumn where the flow of the sample through the microcolumn reduces the $PbO_2$ solid phase reagent to Pb(II), which is measured by flame atomic absorption spectrometry. The absorbance of Pb(II) are proportional to the concentration of nitrite in the samples. The calibration curve was linear up to 30 mg $L ^{-1}$, with a detection limit of 0.11 mg $L ^{-1}$ for a 400 mL injected sample volume and a sampling rate of about 80 $h ^{-1}$. The results exhibit no interference from the presence of large amounts of ions. The developed procedure was found to be suitable for the determination of nitrite in foodstuffs and wastewaters. A relative standard deviation better than 0.9% was obtained in a repeatability study. The reliability of the method was established by parallel determination against the standard method.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2004년도 추계학술대회
• /
• pp.1140-1145
• /
• 2004

Laser-induced incandescence (LII) is introduced as a valuable tool for the characterization of nanoparticles in flame environments. This technique is based on the heating of the particles by a short laser pulse and the subsequent detection of the thermal radiation. It has been applied successfully for the investigation of soot in different fields of application. The evaluation of the temporal decay of the laser-induced incandescence (LII) signal from soot particles is introduced as a technique to obtain two-dimensional distributions of particle sizes and is applied to a laminar diffusion flame. This novel approach to soot sizing exhibits several theoretical and technical advantages compared with the established combination of elastic scattering and LII, especially as it yields absolute sizes of primary particles without requiring calibration. With this technique a spatially resolved 2-D measurement of soot primary particle sizes is feasible in a combination process form the ratio of emission signals obtained at two delay times after a laser pulse, as the cooling behavior is characteristic of particle size.

• 센서학회지
• /
• 제29권2호
• /
• pp.112-117
• /
• 2020

Xylene is a hazardous volatile organic compound that should be precisely measured to monitor indoor air quality. However, the selective and sensitive detection of ppm-level xylene using oxide-semiconductor gas sensors remains a challenge. In this study, pure and Cr-doped Co₃O₄ nanoparticles (NPs) were prepared using flame spray pyrolysis, and their gas-sensing characteristics to 5-ppm xylene at 250 ℃ were investigated. The 4 at% Cr-doped Co₃O₄ NPs exhibited a high gas response to 5-ppm xylene (resistance ratio to gas and air = 39.1) and negligible cross-responses to other representative and ubiquitous indoor pollutants such as ethanol, benzene, formaldehyde, carbon monoxide, and ammonia. In this paper, the enhancement of the gas response and selectivity of Co₃O₄ NPs to xylene by Cr doping was discussed in relation to the catalytic promotion of the gas-sensing reaction. This sensor can be used to monitor indoor xylene.