• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.630-633
• /
• 2020

무의식적인 손-얼굴의 접촉으로 인한 감염의 문제점을 해결하기 위해, 얼굴 만지기 행동을 인식할 필요가 있다. 본 연구는 최근 각광을 받는 딥러닝 기술을 이용하여 비디오 영상에서 얼굴 만지기 행동 인식에 대한 연구이다. 우선, 비디오 영상에서 얼굴 만지기와 관련된 11 가지 행동에 대한 시, 공간적 특징을 컨볼루션 신경망을 통해 추출한다. 추출된 정보는 각 행동 레이블로 인코딩되어 비디오 영상에서 얼굴 만지기 행동을 분류한다. 또한, 3D, 2D 컨볼루션 신경망의 대표 네트워크인 I3D, MobileNet v3에 대해 비교 실험을 진행한다. 제안하는 시스템을 적용하여 인간의 행동을 분류하는 실험을 진행했을 때, 얼굴을 만지는 행동을 99%의 확률로 구분했다. 이 시스템을 이용하여 일반인이 무의식적인 얼굴 만지기 행동에 대해서 정량적으로 또는 적시적으로 인식을 하여, 안전한 위생 습관을 확립하여 감염의 확산방지에 도움을 줄수 있기를 바란다.

• Proceedings of the Korean Society of Disaster Information Conference
• /
• 2023.11a
• /
• pp.73-74
• /
• 2023

본 논문에서는 음향센서를 기반으로 한 지하시설물 화재 위험감지 방법을 제안하였다. 음향센서는 진동이나 광센서처럼 접촉식이 아니기 때문에 결로가 발생하고 있는 취약구간에 설치하여 보다 효율적으로 활용이 가능하고 지하시설물 내부에 설치된 기기나 장비들과 상호작용하거나 간섭하지 않기 때문에 안전하게 관리가 가능하다. 이러한 특징으로 지하 시설물에서 내 통행이 불편하여 관리하기 힘든 구간이나 결로가 많아 화재안전에 주의가 필요한 곳에 설치하여 전기스파크 발생 감지를통해 재난이 발생하기 이전 화재위험을 감지하는 방법론 중 하나가 될 수 있다. 제안하는 방법은 음향 센서를 통해 지하공동구 안에서 발생하는 소리들을 수집하고 일정한 길이의 시간 단위 프레임들로 분할한 후 분석하여 전기스파크의 특징 벡터를 도출한다. 전기스파크 감지 모델로는 전기스파크 신호의 지역적 특성을 포착할 수 있도록 2D-CNN 구조를 사용하며 모델에서 출력된 전기 스파크 발생 예측확률을 분할된 단위 프레임 따라 계산하여 융합한다. 이로 인해 높은 정확도의 전기스파크 감지 정밀도를 얻을 수 있으며, 이는 전기 스파크에 의한 화재 이벤트 감지 있어서 효과적인 센싱 기술임을 알 수 있다.

• Annual Conference of KIPS
• /
• 2023.11a
• /
• pp.1012-1013
• /
• 2023

본 논문의 사회적 취약지구를 위한 전용 화재감지 AI'sTory 플랫폼은 달동네와 판자촌과 같이 소외계층이 집약된 곳이라면 노후된 기기와 전선들이 많을 것이고 그렇다면 전기적 요인으로 인한 화재가 발생할 것이라는 가설에서부터 시작한다. 피해의 최소화를 위해 고안한 사회적 취약지구를 위한 전용 화재감지 AI'sTory 는 스파크를 감지하고, 전선 주위의 발열량을 모니터링하며, 화재를 감지한 즉시 소방서로 연락을 취하는 인공지능 시스템이다. 더 나아가 센서로부터 데이터를 받아 매일 온·습도, 전력 접촉부의 온도, 계절 등 여러 환경을 고려해 화재가 발생할 확률을 계산한다. 또한 화재가 발생하지 않아도 매일 화재 발생 위험률을 제공한다. 이 정보를 제공하기 위해 사회적 취약지구를 위한 전용 화재감지 AI'sTory 는 하드웨어와 소프트웨어, 인공지능으로 구성하였다.

• The KIPS Transactions:PartC
• /
• v.17C no.4
• /
• pp.307-314
• /
• 2010

Recently, it is proved that contactless smart-card based RFID tags, which is used for proximity authentication, are vulnerable to relay attacks with various location-based attacks such as distance fraud, mafia fraud and terrorist fraud attacks. Moreover, distance bounding protocols have been researched to prevent these relay attacks that can measure the message transmitted round-trip time between the reader and the tag. In 2005, Hancke and Kuhn first proposed an RFID distance bounding protocol based on secure hash function. However, the Hancke-Kuhn protocol cannot completely prevent the relay attacks because an adversary has (3/4)$^n$ attack success probability. Thus, this paper proposes a new distance-bounding protocol for light-weight RFID systems that can reduce to (5/8)$^n$ for the adversary's attack success probability. As a result, the proposed protocol not only can provide high-space efficient based on a secure hash function and XOR operation, but also can provide strong security against the relay attacks because the adversary's attack success probability is optimized to (5/8)$^n$.

• The Journal of the Korea Contents Association
• /
• v.16 no.10
• /
• pp.196-205
• /
• 2016

The purpose of this study is to identify factors influencing treatment result in patients with Tuberculosis by patient characteristic, admission and disease characteristic, and hospital characteristic from 2006 to 2012. Survey data was using Korean national hospital discharge in-depth survey data produced by KCDC(Korea Center for Disease Control and Prevention). Study subjects were 8,305 inpatients with TB(A15.0~A19.9) and analyzed frequency, chi-square test, and logistic regression by using SPSS 20(Statistical Package for the Science). The results of this study show that influencing factors of treatment result were ages (20-39, 40-64, and over 65 years), type of insurance(medical aid), disease code (A16, A17, A18, A19), LOS (31-90, and 91-180 days), beds of hospital (300-499, 500-999, over 1,000 beds) and hospital district (non-metropolitan). These findings implied that it is necessary to support successful prevention and management for high risk TB groups and to build middle and long-term policies as well as short -term policy.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.15 no.3
• /
• pp.142-154
• /
• 2011

This study predicted the probability of corrosion initiation of reinforced concrete tunnel boxes structures using the Monte Carlo Simulation. For the inner wall and outer wall in the tunnel boxes, exposed to airborne chloride ion and seawater directly respectively, statistical values of parameters like diffusion coefficient D, surface chloride content $C_s$, cover depth c, and the chloride threshold level $C_{lim}$ were examined from experiment or literature review. Their average values accounted for $3.77{\times}10^{-12}m^2/s$, 3.0% by weight of cement, 94.7mm and 45.5mm for outer wall and inner wall, respectively, and 0.69% by weight of cement for D, $C_s$, c, and $C_{lim}$, respectively. With these parametric values, the distribution of chloride contents at rebar with time and the probability of corrosion initiation of the tunnel boxes, inner wall and outer wall, was examined by considering time dependency of chloride transport. From the examination, the histogram of chloride contents at rebar is closer to a gamma distribution, and the mean value increases with time, while the coefficient of variance decreases with time. It was found that the probability of corrosion initiation and the time to corrosion were dependent on the time dependency of chloride transport. Time independent model predicted time to corrosion initiation of inner wall and outer wall as 8 and 12 years, respectively, while 178 and 283 years of time to corrosion was calculated by time dependent model for inner wall and outer wall, respectively. For time independent model, the probability of corrosion at 100 years of exposure for inner wall and outer wall was ranged 59.5 and 95.5%, respectively, while time dependent model indicated 2.9 and 0.2% of the probability corrosion, respectively. Finally, impact of $C_{lim}$, including values specified in current codes, on the probability of corrosion initiation and corrosion free life is discussed.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.6
• /
• pp.89-98
• /
• 2005

Fingerprint segmentation and ridge orientation estimation algorithms with images from a mobile camera are proposed. The fingerprint images from a mobile camera are quite different from those from conventional sensor, called touch based sensor such as optical, capacitive, and thermal. For example, the images from a mobile camera are colored and the backgrounds or non-finger regions are very erratic depending on how the image capture time and place. Also the contrast between ridge and valley of a mobile camera image are lower than that of touch based sensor image. To segment fingerprint region, we first detect the initial region using color information and texture information. The LUT (Look Up Table) is used to model the color distribution of fingerprint images using manually segmented images and frequency information is extracted to discriminate between in focused fingerprint regions and out of focused background regions. With the detected initial region, the region growing algerian is executed to segment final fingerprint region. In fingerprint orientation estimation, the problem of gradient based method is very sensitive to outlier that occurred by scar and camera noise. To solve this problem, we propose a robust regression method that removes the outlier iteratively and effectively. In the experiments, we evaluated the result of the proposed fingerprint segmentation algerian using 600 manually segmented images and compared the orientation algorithms in terms of recognition accuracy.

• The Journal of Korean Association of Computer Education
• /
• v.16 no.2
• /
• pp.61-68
• /
• 2013

SNS is a social networking service that helps people to have a two-way communication, manage their personal relationships and share information. The domestic and international SNS markets have attained a steady growth, and their growth is being more accelerated recently. Under these circumstances, immature students are more likely to show negative cyber behavior. This study attempted to analyze the relationship between the use of SNS, motives of SNS use, the use of active SNS functions, SNS-dependency and views in SNS and negative SNS behaviors among elementary and middle school students. For this, negative cycber behaviors are classified into four stages depending on the severity, for each of which distribution of factors is investigated and the combination of factors to determine each stage is obtained through association rule analysis. As a result, it is found that 85% of the students rarely show negative cyber behaviors, stealing personal information and contacting with strangers are the most frequent negative behaviors, and students with a great dependency on SNS are highly probable to show negative behaviors.

• Proceedings of the Korean Society of Tribologists and Lubrication Engineers Conference
• /
• 2004.11a
• /
• pp.296-300
• /
• 2004

Currently. the study on structural ceramic helps to suggest the precise data of friction and wear in accordance with the various conditions in operations. Also, the study helps to predict effective operating conditions by monitoring the occurrence of wear transition. The studies in the Past were mainly concentrated in using identical materials. However, it is highly likely to have unqualified data from the differences of mechanical and chemical properties between ceramic materials. Thus, in this study, through conducting the ball-on-disk type wear testing, the different ceramic materials has been used to consider tribological characteristics between different ceramic materials. We conducted the wear test by using three kinds of specimen which are zirconia, alumina and silicon carbide against zirconia. We have changed the sliding velocity and the loading conditions in this test and found out that there is row friction coefficient and wear rate in the combination of zirconia and silicon carbide.

• Membrane Journal
• /
• v.10 no.3
• /
• pp.148-154
• /
• 2000

Poly(vinyl alcohol)(PVA)/sulfur-siccinic acid(SSA) membrane performances have been studied for the vapor permeation separation of methyl tort-butyl ether(MTBE)/methanol mixtures with varying operation temperatures, amount of cross-linking agents, and feed compositions. 1'here are two factors, the membrane network and the hydrogen bonding, in the swelling measurements of PVA/SSA membranes. These two factors act interdependently on the membrane swelling. The sulfuric acid group in SSA took an important role in the membrane performance. The cross-linking effect might be more dominant than the hydrogen bonding effect due to the sulfuric acid group at 7% SSA membrane. Hydrogen bonding effect was more important for 5% SSA membrane. In vapor permeation, density or concentration of methanol in vapor feed is lower than that of methanol in liquid feed, as a result, the hydrogen bonding portion between the solvent and the hydroxyl group in PVA is reduced in vapor permeation. In this case, the 7% SSA membrane shows the highest separation factor of 2187 with the flux of 4.84g/$m^2$hr for MTBE/methanol=80/20 mixtures at 3$0^{\circ}C$.