• 조명전기설비학회논문지
• /
• 제17권4호
• /
• pp.106-115
• /
• 2003

보안 방범설비 시스템 구축시 교차식 스테레오 카메라 방식으로 이동객체를 추적하고, 카메라에서 이동 객체까지 거리를 측정하는 새로운 알고리즘을 제안하였다. 이동 객체를 추출하기 위해서 스테레오 입력 영상을 화소 단위로 정합하였으며, 입력 영상의 주변명암 변화에 대한 오류를 보정하기 위하여 적응형 임계값으로 이동성분을 구한 후 잡음제거 알고리즘으로 이동객체 영역만 추출하였다. 그리고 이동 객체의 위치값을 이용하여 카메라의 팬/틸트를 제어함으로써 이동객체를 추적하였으며, 교차식 스테레오 카메라 시스템의 파라미터 특징을 이용하여 이동객체까지의 거리를 측정하는 방법을 제시하였다. 실험 결과 제안한 알고리즘은 스테레오 방범 카메라 시스템에서 이동객체의 거리 측정과 추적이 가능함을 알 수 있었다. 거리측정 오차도 6%이내에 존재하였다. 따라서 제안한 알고리즘을 이용하여 시스템을 구현할 경우 스테레오 방범 시스템, 이동자율 로봇 시스템 및 스테레오 원격제어 시스템 등에 응용할 수 있을 것이다.

• Korean Chemical Engineering Research
• /
• 제57권2호
• /
• pp.225-231
• /
• 2019

우리나라의 경우 지상에 설치된 고압 천연가스 배관과 건축물 간의 최소 이격거리는 가스기술기준(KGS code)에 의해 규제된다. 이 논문을 통해 이러한 최소 이격거리를 관련된 KGS 코드를 개정한 기술적 근거를 보여주고자 한다. 이격거리를 설정하는 접근 방법으로 합리적 사고 시나리오에 의한 피해기반 접근법을 적용하였는데 배관에 부착된 1인치 분기 라인이 파손되어 제트화재가 발생한 시나리오를 선정하였다. 여기서 공업지역에 종사하는 작업자에 대해 비공업지역에 있는 사람들보다 더 높은 허용가능 복사열 플럭스를 적용하였다. 그 이유는 공업지역에 종사하는 근로자는 일반 대중들 보다 더 짧은 시간 안에 비상 대피가 가능하기 때문이다. 이 사고 시나리오에 대한 피해영향 분석 결과로부터 지상에 설치된 고압 천연가스 배관과 건축물 간의 최소 이격거리로서 비공업지역에서는 30 m, 공업지역에서는 15 m로 KGS 코드 개정을 제안하였다. 코드 개정안은 KGS 코드 위원회(가스기술기준위원회)에 채택되어 현재 시행 중이다.

• 한국정보통신학회논문지
• /
• 제24권4호
• /
• pp.494-500
• /
• 2020

본 논문에서는 초광대역 (Ultra-wideband, UWB) 시스템에서 실내 위치 측위를 위한 새로운 거리 추정 기법을 제안한다. 제안하는 기법은 딥러닝 기법 중 하나인 순환 신경망 (RNN)을 기반으로 한다. 순환신경망은 시계열 신호를 처리하는데 유용한데 UWB 신호 역시 시계열 데이터로 볼 수 있기 때문에 순환신경망을 사용한다. 구체적으로, UWB 신호가 IEEE 802.15.4a 실내 채널모델을 통과하고 수신된 신호에서 순환신경망 회귀를 통해 송신기와 수신기 사이의 거리를 추정하도록 학습한다. 이렇게 학습된 순환신경망 모델의 성능은 새로운 수신신호를 이용하여 검증하며 기존의 임계값 기반의 거리 추정 기법과도 비교한다. 성능지표로는 제곱근 평균추정에러 (root mean square error, RMSE)를 사용한다. 컴퓨터 모의실험 결과에 따르면 제안하는 거리 추정 기법은 수신신호의 신호 대 잡음비 (signal to noise ratio, SNR) 및 송수신기 사이의 거리와 상관없이 기존 기법보다 항상 월등히 우수한 성능을 보인다.

• 제어로봇시스템학회논문지
• /
• 제22권9호
• /
• pp.738-744
• /
• 2016

In this paper, a robust visual odometry system has been proposed and implemented in an environment with dynamic illumination. Visual odometry is based on stereo images to estimate the distance to an object. It is very difficult to realize a highly accurate and stable estimation because image quality is highly dependent on the illumination, which is a major disadvantage of visual odometry. Therefore, in order to solve the problem of low performance during the feature detection phase that is caused by illumination variations, it is suggested to determine an optimal threshold value in the image binarization and to use an adaptive threshold value for feature detection. A feature point direction and a magnitude of the motion vector that is not uniform are utilized as the features. The performance of feature detection has been improved by the RANSAC algorithm. As a result, the position of a mobile robot has been estimated using the feature points. The experimental results demonstrated that the proposed approach has superior performance against illumination variations.

• Journal of information and communication convergence engineering
• /
• 제14권1호
• /
• pp.40-44
• /
• 2016

The device coupling between the stacked top/bottom field-effect transistors (FETs) in two types of monolithic 3D inverter (M3INV) with/without a metal layer in the bottom tier is investigated, and then the regime of the thickness T_ILD and dielectric constant ε_r of the inter-layer distance (ILD), the doping concentration N_d (N_a), and length L_g of the channel, and the side-wall length L_SW where the stacked FETs are coupled are studied. When N_d (N_a) < 10¹⁶ cm^-3 and L_SW < 20 nm, the threshold voltage shift of the top FET varies almost constantly by the gate voltage of the bottom FET, but when N_d (N_a) > 10¹⁶ cm^-3 or L_SW > 20 nm, the shift decreases and increases, respectively. M3INVs with T_ILD ≥ 50 nm and ε_r ≤ 3.9 can neglect the interaction between the stacked FETs, but when T_ILD or ε_r do not meet the above conditions, the interaction must be taken into consideration.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 The Third Asian Fuzzy Systems Symposium
• /
• pp.288-295
• /
• 1998

This paper proposes a new data classification method based on the tolerant rough set that extends the existing equivalent rough set. Similarity measure between two data is described by a distance function of all constituent attributes and they are defined to be tolerant when their similarity measure exceeds a similarity threshold value. The determination of optimal similarity theshold value is very important for the accurate classification. So, we determine it optimally by using the genetic algorithm (GA), where the goal of evolution is to balance two requirements such that (1) some tolerant objects are required to be included in the same class as many as possible. After finding the optimal similarity threshold value, a tolerant set of each object is obtained and the data set is grounded into the lower and upper approximation set depending on the coincidence of their classes. We propose a two-stage classification method that all data are classified by using the lower approxi ation at the first stage and then the non-classified data at the first stage are classified again by using the rough membership functions obtained from the upper approximation set. We apply the proposed classification method to the handwritten numeral character classification. problem and compare its classification performance and learning time with those of the feed forward neural network's back propagation algorithm.

• 한국군사과학기술학회지
• /
• 제14권1호
• /
• pp.123-131
• /
• 2011

In this paper, modeling and design of a distributed detection system are considered for an active sonar sensor network. The sensor network has a parallel configuration and it consists of a fusion center and a set of receiver nodes. A system with two receiver nodes is considered to investigate a theoretical aspect of design. To be specific, AND rule and OR rule are considered as the fusion rules of the sensor network. For the fusion rules, it is shown that a threshold rule of each sensor node has uniformly most powerful properties. Optimum threshold for each sensor is obtained that maximizes the probability of detection given probability of false alarm. Numerical experiments were also performed to investigate the detection characteristics of a distributed detection system with multiple sensor nodes. The experimental results show how signal strength, false alarm probability, and the distance between nodes in a sensor field affect the system detection performances.

• 대한조선학회논문집
• /
• 제43권4호
• /
• pp.494-503
• /
• 2006

Inherent damage zone model is presented to explain the fatigue properties near the fatigue limit and the crack growth threshold consistently Inherent damage zone model assumes that the stress at a point which is located at a small distance, $r_0$, an inherent length of the material that represents the size of effective damage zone, from the crack initiation position governs the fatigue characteristics regardless of the geometric configuration of the specimen; smooth specimen, notched specimen or cracked specimens with short and long crack length. A special feature of the paper is using the exact stress distributions of notched and cracked specimens at the strength evaluations. Analytical elastic solutions by Neuber and Westergaard are employed for this purpose Relationship between fatigue limit of smooth specimen and threshold stress of cracked specimen, occurrence condition of non-propagating crack at the root of elliptic notch and circular hole and relationship between stress concentration factor and fatigue notch factor are discussed quantitatively based on the proposed model.

• 한국군사과학기술학회지
• /
• 제12권3호
• /
• pp.333-343
• /
• 2009

Tactical Information Communication Network(TICN) is a concept-type integrated Military Communication system that enables precise command control and decision making by unifying the diversified military communication network and conveying diverse range of battle field information on real-time, at right place at right time. TICN is designed to advance into high speed, large capacity, long distance wireless relay transmission. To support mobility in battlefield environments, the application of Ad-hoc networking technology to its wireless communication has been examined. Ad-hoc network is consist of mobile nodes and nodes in the network depends on the cooperation of other nodes for forwarding of packets. In this context, some non-cooperating nodes may delay forwarding of packets or drop the packets. This may hamper the network as a whole and disrupt communication between the cooperating nodes. To solve this problem, we present a solution with a Node Weight Management Server(NWMS), which manages each node's weight according to its behavior in local area. When the NWMS detects misbehaving node, it increases the node's weight. If the node's weight exceeds a predefined threshold then the NWMS broadcasts the node's information into network to isolate the misbehaving node from the network. These mechanisms show that they are highly effective and can reliably detect a multitude of misbehaving node.

• 한국군사과학기술학회지
• /
• 제12권2호
• /
• pp.159-169
• /
• 2009

In this paper, optimum design of distributed detection is considered for a parallel sensor network system consisting of a fusion center and two passive sonar nodes. AND rule and OR rule are employed as the fusion rules of the sensor network. For the fusion rules, it is shown that a threshold rule of each sensor node has uniformly most powerful properties. Optimum threshold for each sensor is investigated that maximizes the probability of detection under the constraint of a specified probability of false alarm. It is also investigated through numerical experiments how signal strength, false alarm probability, and the distance between two sensor nodes affect the system detection performances.