• 한국정보과학회논문지:소프트웨어및응용
• /
• 제31권4호
• /
• pp.403-410
• /
• 2004

바이모달 음성인식기는 잡음 환경하 음성인식 성능을 향상하기 위해 고안되었다. 바이모달 음 성인식기에 있어 영상을 통한 시각 특징 추출은 매우 중요한 역할을 하며 이를 위한 입술 위치 검출은 시각 특징 추출을 위한 중요한 선결 과제이다 본 논문은 색상분포와 SVM을 이용하여 시각 특징 추출을 위한 입술 위치 검출 방법을 제안하였다. 제안된 방법은 얼굴색/입술 색상 분포를 학습하여 이로부터 입술의 초기 위치를 빠르게 찾아내고 SVM을 이용하여 입술의 정확한 위치를 찾음으로써 정확하고 빠르게 입술의 위치를 찾도록 하였으며 실험을 통해 바이모달 인식기에 적용하기에 적합함을 알 수 있었다.

• Journal of Power Electronics
• /
• 제10권2호
• /
• pp.155-164
• /
• 2010

This paper presents a voltage control algorithm for a hybrid multilevel inverter based on a staged-perception of the inverter voltage vector diagram. The algorithm is applied to control a three-stage eighteen-level hybrid inverter, which has been designed with a maximum number of symmetrical levels. The inverter has a two-level main stage built using a conventional six-switch inverter and medium- and low- voltage three-level stages constructed using cascaded H-bridge cells. The distinctive feature of the proposed algorithm is its ability to avoid the undesirable high switching frequency for high- and medium- voltage stages despite the fact that the inverter's dc sources voltages are selected to maximize the number of levels by state redundancy elimination. The high- and medium- voltage stages switching algorithms have been developed to assure fundamental switching frequency operation of the high voltage stage and not more than few times this frequency for the medium voltage stage. The low voltage stage is controlled using a SVPWM to achieve the reference voltage vector exactly and to set the order of the dominant harmonics. The inverter has been constructed and the control algorithm has been implemented. Test results show that the proposed algorithm achieves the desired features and all of the major hypotheses have been verified.

• Journal of Electrical Engineering and Technology
• /
• 제9권3호
• /
• pp.908-925
• /
• 2014

Induction Motor (IM) has several desirable features for high performance adjustablespeed operation. This paper presents the design of a robust controller for vector control induction motor drive performances improvement. Proposed predictive speed controller, which is aimed to guarantee the stability of the closed loop, is based on the Poisson-Laguerre (PL) models for the association vector control drive and the induction motor; without necessity of any mechanical parameter, and requires only two control parameters to ensure implicitly the integrator effect on the steady state error, load torque disturbances rejection and anti-windup effect. In order to improve robustness, insensitivity against external disturbances and preserve desired performance, adaptive control is added with the aim to ensure an online identification of controller parameters through an online PL models identification. The proposed control is compared with the conventional approach using PI controller. Simulation with MATLAB/SIMULINK software and experimental results for a 1kW induction motor using a dSPACE system with DS1104 controller board are carried out to show the improvement performance.

• Journal of Power Electronics
• /
• 제3권4호
• /
• pp.239-248
• /
• 2003

This paper presents a novel prototype of three-phase voltage source type zero voltage soft-switching inverter with the auxiliary resonant snubbers suitable for high-power applications with IGBT power module packages in order to reduce their switching power losses as well as electromagnetic conductive and radiative noises. A proposed single inductor-assisted resonant AC link snubber circuit topology as one of some auxiliary resonant commutation snubbers developed previously to achieve the zero voltage soft-switching (ZVS) for the three-phase voltage source type sinewave PWM inverter operating under the instantaneous space voltage vector modulation is originally demonstrated as compared with the other types of resonant AC link snubber circuit topologies. In addition to this, its operation principle and unique features are described in this paper. Furthermore, the practical basic operating performances of the new conceptual instantaneous space voltage vector modulation resonant AC link snubber-assisted three-phase voltage source type soft-switching PWM inverter using IGBT power module packages are evaluated and discussed on the basis of switching voltage and current waveforms, output line to line voltage quality, power loss analysis, actual power conversion efficiency and electromagnetic conductive and radiative noises from an experimental point of view, comparing with those of conventional three-phase voltage source hard-switching PWM inverter using IGBT power modules.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2002년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.517-517
• /
• 2002

Recently in the field of telecommunication, there is much interest in geo-surface characteristics of urban areas. Geophysical properties of urban features are now incorporated with accurate positional information to model the telecommunication environment. In this study, three-dimensional buildings are geometrically reconstructed from existing vector maps and aerial images. Accurate digital vector maps are easily available in Korea. However existing maps, which had been produced for GIS applications, do not have height information which is critical to three dimensional building reconstruction. Image matching techniques were applied to aerial image stereopairs to automatically extract the height information of buildings. Planimetric coordinates from vector maps were used as initial guides in the process. Future studies will be undertaken to link geophysical properties to the three-dimensional spatial objects reconstructed from this study thus bringing the telecommunication environment model closer to reality.

• 한국전자통신학회논문지
• /
• 제14권2호
• /
• pp.275-282
• /
• 2019

산이나 건물과 같은 지형지물에 의해 발생하는 클러터는 대기 상태와 무관한 오염된 자료로 간주된다. 클러터 신호를 줄이기 위한 기본적인 방법은 윈드프로파일러 주위에 클러터 펜스를 설치하는 것이다. 클러터 펜스가 윈드프로파일러 관측 자료에 미치는 효과를 조사하기 위해 클러터 펜스 설치 전과 후의 윈드프로파일러 바람 벡터 수집률과 바람 벡터의 정확도를 고도별로 조사하였다. 윈드프로파일러의 클러터 펜스는 자료 수집률 뿐만 아니라 자료의 품질을 향상시키는데 기여하였다.

• 한국컴퓨터정보학회논문지
• /
• 제27권7호
• /
• pp.27-33
• /
• 2022

본 논문은 콘텐츠 이미지에 스타일 이미지를 입혀 스타일이 적용된 이미지를 생성하고자 기존의 이미지 대 이미지 방법이 아닌 다중 도메인 스타일 트랜스퍼(style transfer) 방법을 적용한다. 도메인별로 데이터 분포에 대한 다양한 특성을 고려하고, 콘텐츠 데이터의 특징을 잘 보존하면서 높은 품질의 이미지가 생성되도록 잠재 벡터(latent vector)와 가우시안 노이즈를 추가하는 새로운 GAN의 생성자(generator) 아키텍처를 제안한다. 콘텐츠 이미지가 도메인별 스타일을 잘 학습할 수 있도록 네트워크를 구성하고 사계절 이미지로 구성된 도메인에 적용하여 고해상도의 스타일 트랜스퍼 결과를 보여준다.

• Nuclear Engineering and Technology
• /
• 제55권3호
• /
• pp.827-838
• /
• 2023

Centrifugal pump is a key part of nuclear power plant systems, and its health status is critical to the safety and reliability of nuclear power plants. Therefore, fault diagnosis is required for centrifugal pump. Traditional fault diagnosis methods have difficulty extracting fault features from nonlinear and non-stationary signals, resulting in low diagnostic accuracy. In this paper, a new fault diagnosis method is proposed based on the improved particle swarm optimization (IPSO) algorithm-based variational modal decomposition (VMD) and relevance vector machine (RVM). Firstly, a simulation test bench for rotor faults is built, in which vibration displacement signals of the rotor are also collected by eddy current sensors. Then, the improved particle swarm algorithm is used to optimize the VMD to achieve adaptive decomposition of vibration displacement signals. Meanwhile, a screening criterion based on the minimum Kullback-Leibler (K-L) divergence value is established to extract the primary intrinsic modal function (IMF) component. Eventually, the factors are obtained from the primary IMF component to form a fault feature vector, and fault patterns are recognized using the RVM model. The results show that the extraction of the fault information and fault diagnosis classification have been improved, and the average accuracy could reach 97.87%.

• 한국음향학회지
• /
• 제28권5호
• /
• pp.471-476
• /
• 2009

본 논문에서는 변별적 가중치 학습 (discriminative weight training) 기반의 최적화된 가중치를 가지는 입력벡터를 구성하여 support vector machine (SVM)을 이용한 기존의 3GPP2 selectable mode vocoder (SMV)코덱의 음성/음악 분류 성능을 향상 시키는 방법을 제안한다. 구체적으로, 최소 분류 오차 minimum classification error (MCE) 방법을 도입하여, 최적화된 가중치를 각각의 특징벡터별로 부가한 SVM을 적용하여 기존의 가중치를 고려하지 않은 SVM 기반의 알고리즘과 비교하였으며, 우수한 음성/음악 분류 성능을 보였다.

• 정보보호학회논문지
• /
• 제24권1호
• /
• pp.87-95
• /
• 2014

인가된 AP(Access Point)에 대해서만 사용을 허용하는 무선 통신 환경에서, 스마트폰의 테더링(tethering) 기능에 의한 로그 AP(rogue AP) 사용은 자료 유출과 같은 심각한 보안 문제를 발생시킨다. 이에 본 논문은 홉(hop) 간 RTT 값을 특징점(Feature)으로 하고 분류기로 k-SVM(Kernel Support Vector Machine)를 사용하여 로그 AP를 탐지하는 방법을 제안한다. 실험을 통해 유선 네트워크를 이용하여 설치한 일반 AP와 LTE 망을 활용하여 설치된 로그 AP 간의 구분이 가능함을 보였다.