• Journal of Advanced Marine Engineering and Technology
• /
• 제36권5호
• /
• pp.645-661
• /
• 2012

Recently, pirates are infesting on the sea and they have been hijacking the several vessels for example Samho Dream and Samho Jewelry of Korea. One of the items to reduce the risk is to adopt the invader detection system. If the pirates break in to the ship, the detection system can monitor the pirates and then call the security alarm. The crew can gain time to hide to the safe room and the report can be automatically sent to the control room to cope with the situation. For the invader detection, an unmanned observation system was proposed using the image detection algorithm that extracts the invader image from the recording image. To detect the motion area, the difference value was calculated between the current image and the prior image of the invader, and the 'AND' operator was used in calculated image and edge line. The image noise was reduced based on the morphology operation and then the image was transformed into morphological information. Finally, a neural network model was applied to recognize the invader. In the experimental results, it was confirmed that the proposed approach can improve the performance of the recognition in the invader monitoring system.

• 전기전자학회논문지
• /
• 제25권2호
• /
• pp.330-336
• /
• 2021

하드웨어 이중화 구성 수를 줄이는 대표적인 방법은 마이크로컨트롤러로 고장을 검출, 식별 및 수용을 위한 해석적 기법으로 구현하는 것이다. 본 논문에서는 해석적 기법 중 하나인 고장 검출 필터를 항공기 터보팬 엔진 시스템에 적용하였다. 고장 검출 필터는 특수한 형태의 관측기로써 특정한 고장 발생시 잔차가 출력 공간에서 일정한 방향을 유지함으로써 고장의 위치 판별이 가능한 장점이 있다. 이에 본 논문에서는 터보팬 엔진 내 공기 터빈 시스템의 단일 입출력 동적 시스템 모델링, 고장 검출 필터 설계 및 이를 적용한 모의실험 결과를 나타내었다. 모의실험 결과를 통해 고장 검출 필터가 갖는 방향성에 대한 민감성 효과로 고장 검출이 유효하게 적용될 수 있음을 보였다.

• 대한전기학회논문지:전기물성ㆍ응용부문C
• /
• 제55권11호
• /
• pp.532-537
• /
• 2006

Lightning induced faults accounts for more than 66% at the transmission lines of KEPCO. The lightning causes damages to power system equipments including transmission line, the blackout of electricity and the electro-magnetic interference. Because of this reason, we need the real time lightning information for the optimal operation of power system. And, it is required to obtain and accumulate the lightning current parameters for the insulation design. In 2005, KEPRI constructed a lightning detection network, the KLDNet (i.e. Kepco Lightning Detection & Information System) and launched a lightning information service for KEPCO customers. It is intended to provide data service on the operation of transmission lines and collect lightning-related data, which is the most important factor regulating power system design and operation. The new system will replace LPATS, the old detection system, which has been operating since 1995 and is rapidly failing in terms of both detection performance and location accuracy. The purpose of this paper is to explain the work performed and the results of that work in performing a site survey of several locations. The purpose of the site survey is to find locations acceptable for the installation of a lightning location receiver in support of a Lightning detection system(LDS). A restriction was placed on the surveyed locations, as they must belong to the Korea Electric Power Company. This requirement was made to facilitate the communication needs of the LDS network. Total of 15 sites were evaluated as possible LDS sensor sites. Some of the sites were rejected for physical reasons and therefore no electrical testing was performed. Of the 15 sites, total of 10 sites were considered acceptable and 5 sites were rejected for various reason. In this paper, we would like to explain the site survey and detection efficiency for LDS.

• 융합보안논문지
• /
• 제17권1호
• /
• pp.31-38
• /
• 2017

네트워크를 통한 사이버 공격 기법들이 다양화, 고급화 되면서 간단한 규칙 기반의 침입 탐지/방지 시스템으로는 지능형 지속 위협(Advanced Persistent Threat: APT) 공격과 같은 새로운 형태의 공격을 찾아내기가 어렵다. 기존에 알려지지 않은 형태의 공격 방식을 탐지하는 이상행위 탐지(anomaly detection)를 위한 해결책으로 최근 기계학습 기법을 침입탐지 시스템에 도입한 연구들이 많다. 기계학습을 이용하는 경우, 사용하는 특징 집합에 침입탐지 시스템의 효율성과 성능이 크게 좌우된다. 일반적으로, 사용하는 특징이 많을수록 침입탐지 시스템의 정확성은 높아지는 반면 탐지를 위해 소요되는 시간이 많아져 긴급성을 요하는 경우 문제가 된다. 논문은 이러한 두 가지 조건을 동시에 충족하는 특징 집합을 찾고자 다목적 유전자 알고리즘을 제안하고 인공신경망에 기반한 네트워크 침입탐지 시스템을 설계한다. 제안한 방법의 성능 평가를 위해 NSL_KDD 데이터를 대상으로 이전에 제안된 방법들과 비교한다.

• 한국멀티미디어학회논문지
• /
• 제13권2호
• /
• pp.171-178
• /
• 2010

무인감시시스템은 무선 칩 같은 기초적인 센서를 이용하는 분야는 많이 연구 되어 왔으며. 카메라를 주요 센서로 하는 영상감시체계 연구 분야가 활성화 되고 있다. 본 논문에서는 다양한 영상검지기법을 조사 분석한 결과를 토대로 영상 객체 인식 기법을 적용한 지능형 영상검지 시스템을 제안하였다. 이 지능형 영상검지 시스템은 사건 전후의 상황을 쉽게 추적 판단 할 수 있으며, 확실한 증거와 다양한 정보를 확보 할 수 있다. 따라서 본 논문에서 제안하는 지능형 영상 검지 시스템은 교통상황 관리, 재난 경보 등 다양한 무인감시시스템에 활용 될 것이다.

• 대한의용생체공학회:의공학회지
• /
• 제26권3호
• /
• pp.145-150
• /
• 2005

A new gait detection system using both FSR (force sensing resistor) sensors and a gyrosensor was developed to detect various gait patterns. FSR sensors were put in self-designed shoe insoles and a gyrosensor was attached to the heel of a shoe. An algorithm was also developed to determine eight different gait transitions during four gait phases: heel-strike, foot-flat, heel-off and swing. The developed system was evaluated from nine heathy mans and twelve hemiplegic patients. Healthy volunteers were asked to walk in various gait patterns: level walking, fore-foot walking and stair walking. Only the level walking was performed in hemiplegic patients. The gait detection system was compared with a optical motion analysis system and the outputs of the FSR sensors. In healthy subjects, the developed system detected successfully more than $99\%$ for both level walking and fore-foot walking. For stair walking, the successful detection rate of the system was above$97\%$. In hemiplegic patients, the developed system detected approximately 98% of gait transitions. The developed gait phase detection system will be helpful not only to determine pathological gait phases but also to apply prosthetics, orthotics and functional electrical stimulation for patients with various gait disorders.

• 융합보안논문지
• /
• 제8권3호
• /
• pp.1-8
• /
• 2008

오용 탐지(misuse-detection) 기반의 침입 탐지 시스템에서 패턴 비교 부분은 많은 수행시간 및 자원을 필요로 하며, 전체 시스템 성능 상의 병목 현상을 유발한다. 더 나아가 이것은 서비스-거부-공격의 표적이 될 여지도 있다. 본 논문에서는 첫째, 서비스 거부 공격에 강한 침입탐지시스템 구성 방안을 살펴보고, 둘째, 특별히 웹서버 시스템에서 효율적으로 탐지를 수행하는 방안을 제안한다. 이 두 가지 방안을 함께 이용함으로써 웹서버 시스템을 서비스 거부 공격을 포함한 침입에 대항하여 효율적으로 안전하게 유지할 수 있다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.974-977
• /
• 2003

The GPS/INS integration system navigation can provide improved navigation performance and has been widely used as a main navigation system for military and commercial vehicles. When two navigation systems are tightly coupled and the structure is complicated, a fault in either the GPS or the INS can lead to a disastrous failure of the whole integration system. This paper proposes a real-time fault detection method for an AGPS/INS integration system. The proposed fault detection method comprises a BIT and a fault detection algorithm based on chi-square test. It is implemented by real-time software modules to apply the AGPS/INS integration system and van test is carried out to evaluate its performance.

• 한국데이타베이스학회:학술대회논문집
• /
• 한국데이타베이스학회 1999년도 춘계공동학술대회: 지식경영과 지식공학
• /
• pp.307-313
• /
• 1999

This paper reviews and assesses the analogy between the human immune system and network intrusion detection systems. The promising results from a growing number of proposed computer immune models for intrusion detection motivate this work. The paper begins by briefly introducing existing intrusion detection systems (IDS's). A set of general requirements for network-based IDS's and the design goals to satisfy these requirements are identified by a careful examination of the literature. An overview of the human immune system is presented and its salient features that can contribute to the design of competent network-based IDS's are analysed. The analysis shows that the coordinated actions of several sophisticated mechanisms of the human immune system satisfy all the identified design goals. Consequently, the paper concludes that the design of a novel network-based IDS based on the human immune system is promising for future network-based IDS's

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2000년도 추계종합학술대회
• /
• pp.665-668
• /
• 2000

This paper presents development of a different substance monitoring system base reliable detection between the conveyer belt and hopper used lot materials transport line of steel company. Conventional detection method of a piece of iron separation system is losed the confidence, because of the place with bad surroundings of measurement so much that materials Production line are completely exposed to dust, moisture and vibration. For the solution of this problem, we developed a different substance detection system using the acoustic emittion sensor and one chip microprocessor which is available for bad surroundings and inexpensive. The reliability of the system was estimated by experiment.