• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.1
• /
• pp.65-70
• /
• 2023

In this paper, we used a deep learning model to detect and segment flame and smoke in real time from fires. To this end, well known U-NET was used to separate and divide the flame and smoke of the fire using multi-class. As a result of learning using the proposed technique, the values of loss error and accuracy are very good at 0.0486 and 0.97996, respectively. The IOU value used in object detection is also very good at 0.849. As a result of predicting fire images that were not used for learning using the learned model, the flame and smoke of fire are well detected and segmented, and smoke color were well distinguished. Proposed method can be used to build fire prediction and detection system.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.499-500
• /
• 2021

현재 4차 산업혁명의 시대로 들어서고 수많은 분야에서 사고와 안전에 관련하여 발전을 이루고 있지만, 매년 화재피해로 인한 인명손실과 재산피해는 다른 사고보다 피해와 규모가 다르며 여전히 고민해야할 숙제이다. 화재 골든타임은 화재 발생직후 5분이내의 시간을 일컫으며, 이 시간을 지키느냐 마느냐에 따라서 화재로 인한 사망자 수는 2배이상 차이가 난다. 본 논문의 화재 감지 시스템은 화재 골든타임을 지키는것에 중점을 두고 이 시스템 자체로 화재를 방지하고, 감지하여 초동조치까지 이루어질 수 있도록 화재 위험요소 차단, 화재 발생시 즉각적인 대처, 인명 피해 최소화, 2차피해 방지를 목적으로 구성되었다. 이로 인해 결과적으로 화재 골든타임을 지켜 화재를 방치시키는 상황이 줄어들 것이고, 인명피해와 재산피해까지 줄일 수 있을것이라 예상된다.

• Journal of Sensor Science and Technology
• /
• v.11 no.4
• /
• pp.209-217
• /
• 2002

The developing of buried fiber optic sensor with high sensitivity and broad-area detecting intruders is carried out using fiber optic ROTDR(Rayleigh Optical Time Domain Reflectometry). The sensing part was designed to be able to broad-area detect intrusion effect per optical fiber length under ground. The bending light losses in optical fibers are investigated by commercial mini ROTDR with wavelength $1.55{\mu}m$, distance range 5km, pulse width 20ns, SNR=5.7. The sensing fibers are selected as the common telecommunication fibers are the 1.5mm, 3.5 mm outer diameter, 4km each length fiber products. Experiments were investigate the characteristics of signal sensitivity according to applied intrusion weight. The relation between the applied weight and the bending loss was almost linear, and broad-area detect intrusion effects are the 2m resolution and $1.3m^2$ per optical fiber length respectively. The light loss by the applied weight on fiber was 0.17 dB/kg. that the sensitivity of the optical fiber sensor was sufficient to detect intruders passing over the buried optical fiber.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.4B
• /
• pp.288-297
• /
• 2003

The fast retransmit and fast recovery algorithm of TCP Reno, when multiple packets in the same window are lost, cannot recover them without RTO (Retransmission Timeout). TCP New-Reno can recover multiple lost packets by extending fast recovery using partial acknowledgement. If the retransmitted packet is lost again during fast recovery, however, RTO cannot be avoided. In this paper, we propose an algorithm called "Duplicate Acknowledgement Counting(DAC)" to alleviate this problem. DAC can detect the retransmitted packet loss by counting duplicate ACKs. Conditions that a lost packet can be recovered by loss recovery of TCP Reno, TCP New-Reno and TCP New-Reno using DAC are derived by modeling loss recovery behavior of each TCP. We calculate the loss recovery probability for random packet loss probability numerically, and show that DAC can improve loss recovery behavior of TCP New-Reno.

• Journal of Energy Engineering
• /
• v.13 no.3
• /
• pp.173-180
• /
• 2004

A technique of the heat rate allocation was devised to monitor the performance of Combined Cycle Power Plant. This calculates the expected heat rate of current conditions and compares it with actual values. Loss allocation in heat rate is reconciled by calculating the magnitude of the deficiency contributed by major components, such as the gas turbine, heat recovery steam generator (HRSG), steam turbine and condenser. Expected power output is determined by a detailed model and correction curves of the plant.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.479-483
• /
• 2012

최근 국내외에서 기상이변에 따른 물 부족에 대응하고 먹는 물의 생산과 공급 효율성 향상을 위하여 스마트워터그리드에 대한 연구가 활발히 진행되고 있는 경향으로 이는 상수도 인프라시설의 운영오류, 자연재해, 사이버를 통한 의도적 공격 등에 대해 안전하고 신뢰성이 높은 시스템을 구축하기 위한 것이다. 또한 상수도 분야에서는 스마트 미터링 개념을 도입하여 상수관망에서 발생되는 각종 사고나 물 손실을 저감하기 위한 노력이 시도되고 있는 실정이다. 일반적으로 누수량이 많을 경우에는 누수의 징후가 지표면에서 확인이 가능하나 미세한 경우 탐사장비의 운영이나 인력의 투입을 통해 가능하다. 물 공급계통인 상수관망에서 물 손실을 저감시키기 위한 방법중 하나로서 수도미터로부터 기록되는 물 사용량 데이터를 이용하여 누수패턴을 추출함으로써 누수감시가 가능하도록 하는 모형을 개발하고자 하였다. 이는 탐사장비에 의한 누수감지와 상호 보완적을 활용할 수 있는 것으로서 사용량 자료를 분석하여 사전에 처리된 자료의 노이즈와 결함 있는 계측값을 필터링시키는 기법이 도입된 것이며 신속한 감지를 통해 누수 지속시간을 감소시킴으로서 누수량의 저감을 목표로 한다. 물 공급 네트워크를 보다 더 효율적 만들 수 있는 누수 감시모형은 상수관망의 운영에 필요한 정보를 도출하기 위하여 보다 정확하고 다양한 수학적 및 통계적 분석을 기반으로 구성되어 누수 외에도 각종 결함을 찾아내는 역할을 수행할 수 있다. 향후 다양한 지역을 이러한 수용가의 물 사용량 미터링 결과를 토대로 실제 사용량과 누수량을 분리하여 분석함으로써 국내 실정에 적합한 누수감시 기술배발 토대를 마련할 수 있을 것으로 판단되며, 누수저감을 위한 실질적인 상수관망의 운영관리 효율향상의 정보로서 활용이 가능할 것으로 판단된다.

• Journal of KIISE:Information Networking
• /
• v.31 no.3
• /
• pp.327-334
• /
• 2004

In this paper, we make MAC protocol improvements for performance enhancement of multi-hop ad-hoc wireless networks. A node in ad-hoc wireless networks can transmit a packet only when the medium is available, and while a packet is being transmitted, no other nodes are allowed to transmit a packet if they are in carrier sensing range. Carrier sensing range can be divided into two disjoint areas of transmission range and carrier sensing zone(9), and we address the importance of the protocol behavior when a node is in carrier sensing zone. The characteristic of the carrier sensing zone is that a node can not know when the remaining time of the on-going transmission session expires or exactly when the media becomes available. Current MAC protocol does not behave in much different way between when a node is in transmission range and in carrier sensing zone. We have conducted a comprehensive simulation to study the performance improvements. The simulation results indicate that the performance is increased and the number of dropped packets due to collision is significantly reduced as much as a half.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.9A
• /
• pp.719-726
• /
• 2009

CR systems performs spectrum sensing operation to detect the appearance of primary users. However, since it is not feasible to do spectrum sensing and data transmission simultaneously, they typically operate alternatively in a time domain. There have been an effort(8) to investigate the optimal spectrum sensing duration for maximum throughput for the scheme with cooperative spectrum sensing. This is based on an assumption that the communication channels between each secondary user and the fusion center are ideal and does not consider the effects of transmission error. Motivated by this, we here model the channels as binary symmetric channels and examined its effect on the maximum throughput and the associated optimal sensing duration. Analysis shows that the performance degradation due to the transmission error is smaller for the case of using the AND fusion rule than for the OR fusion rule.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.7C
• /
• pp.642-650
• /
• 2010

Collaborative spectrum sensing allows secondary users scattered in location to work together to detect the activity of primary users and has been shown to significantly reduce the performance degradation due to fading phenomenon. Most previous works on collaborative spectrum sensing are based on the assumption that local spectrum sensing decisions of secondary users are statistically independent. However, it may not hold in some practical situations with shadowing effect. In this paper, we consider the case that the secondary users are evenly spaced in the form of a linear array and only adjacent secondary users are statistically correlated, and analyze the effect of the statistical correlation on the performance of collaborative spectrum sensing and the throughput of a cognitive radio network. Here we assumed the AND and OR fusion rules for combining the local decisions of secondary users. The analysis showed that the AND fusion rule achieves higher throughput than the OR fusion rule.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.15 no.6
• /
• pp.7-14
• /
• 2019

The automatic monitoring and detection of crowd behavior in the surveillance videos has obtained significant attention in the field of computer vision due to its vast applications such as security, safety and protection of assets etc. Also, the field of crowd analysis is growing upwards in the research community. For this purpose, it is very necessary to detect and analyze the crowd behavior. In this paper, we proposed a deep learning-based method which detects abnormal activities in surveillance cameras installed in a smart city. A fine-tuned VGG-16 model is trained on publicly available benchmark crowd dataset and is tested on real-time streaming. The CCTV camera captures the video stream, when abnormal activity is detected, an alert is generated and is sent to the nearest police station to take immediate action before further loss. We experimentally have proven that the proposed method outperforms over the existing state-of-the-art techniques.