• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1478-1481
• /
• 2005

As the model of most practical system cannot be obtained, the practice of typical control method is limited. Accordingly, numerous artificial intelligence control methods have been used widely. Fuzzy control and neural network control have been an important point in the developing process of the field. This paper is proposed adaptive fuzzy-neural network based on the vector controlled interior permanent magnet synchronous motor drive system. The fuzzy-neural network is first utilized for the speed control. A model reference adaptive scheme is then proposed in which the adaptation mechanism is executed using fuzzy-neural network. Also, this paper is proposed estimation of speed of interior permanent magnet synchronous motor using artificial neural network controller. The back-propagation neural network technique is used to provide a real time adaptive estimation of the motor speed. The error between the desired state variable and the actual one is back-propagated to adjust the rotor speed, so that the actual state variable will coincide with the desired one. The back-propagation mechanism is easy to derive and the estimated speed tracks precisely the actual motor speed. This paper is proposed the analysis results to verify the effectiveness of the new method.

• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.1057-1062
• /
• 2007

얼굴표정 인식은 심리학 연구, 얼굴 애니메이션 합성, 로봇공학, HCI(Human Computer Interaction) 등 다양한 분야에서 중요성이 증가하고 있다. 얼굴표정은 사람의 감정 표현, 관심의 정도와 같은 사회적 상호작용에 있어서 중요한 정보를 제공한다. 얼굴표정 인식은 크게 정지영상을 이용한 방법과 동영상을 이용한 방법으로 나눌 수 있다. 정지영상을 이용할 경우에는 처리량이 적어 속도가 빠르다는 장점이 있지만 얼굴의 변화가 클 경우 매칭, 정합에 의한 인식이 어렵다는 단점이 있다. 동영상을 이용한 얼굴표정 인식 방법은 신경망, Optical Flow, HMM(Hidden Markov Models) 등의 방법을 이용하여 사용자의 표정 변화를 연속적으로 처리할 수 있어 실시간으로 컴퓨터와의 상호작용에 유용하다. 그러나 정지영상에 비해 처리량이 많고 학습이나 데이터베이스 구축을 위한 많은 데이터가 필요하다는 단점이 있다. 본 논문에서 제안하는 실시간 얼굴표정 인식 시스템은 얼굴영역 검출, 얼굴 특징 검출, 얼굴표정 분류, 아바타 제어의 네 가지 과정으로 구성된다. 웹캠을 통하여 입력된 얼굴영상에 대하여 정확한 얼굴영역을 검출하기 위하여 히스토그램 평활화와 참조 화이트(Reference White) 기법을 적용, HT 컬러모델과 PCA(Principle Component Analysis) 변환을 이용하여 얼굴영역을 검출한다. 검출된 얼굴영역에서 얼굴의 기하학적 정보를 이용하여 얼굴의 특징요소의 후보영역을 결정하고 각 특징점들에 대한 템플릿 매칭과 에지를 검출하여 얼굴표정 인식에 필요한 특징을 추출한다. 각각의 검출된 특징점들에 대하여 Optical Flow알고리즘을 적용한 움직임 정보로부터 특징 벡터를 획득한다. 이렇게 획득한 특징 벡터를 SVM(Support Vector Machine)을 이용하여 얼굴표정을 분류하였으며 추출된 얼굴의 특징에 의하여 인식된 얼굴표정을 아바타로 표현하였다.

• The Journal of Korean Association of Computer Education
• /
• v.13 no.6
• /
• pp.79-89
• /
• 2010

In this paper, we proposed a method to find out similar sentences from documents to detect plagiarized documents. The proposed model adapts LSA and N-gram techniques to detect every type of Korean plagiarized sentence type. To evaluate the performance of the model, we constructed experimental data using students' essays on the same theme. Students made their essay by intentionally plagiarizing some reference documents. The experimental results showed that our proposed model outperforms the conventional N-gram model, Vector model, LSA model in precision, recall, and F measures.

• The Transactions of the Korean Institute of Electrical Engineers B
• /
• v.49 no.9
• /
• pp.618-625
• /
• 2000

Recently PWM inverters are widely utilized for many industrial applications e.g. high performance motor drive and PWM techniques are newly developed for an accurate output voltage. Among them space voltage vector PWM(SVPWM) inverter has high voltage ratio and low harmonics compared to the conventional sinusoidal PWM inverter. However output voltage of PWM inverter is distorted and has error duet o the conducting voltage drop of switching devices and the dead time that is inevitable to prevent the shoot-through phenomenon. This paper investigates 3-phase SVPWM inverter which has a new compensation method against dead time and voltage drop. Proposed algorithm calculates gate pulse periods which directly compensates the dead time and nonlinear voltage drop without modification of reference voltages. Direct compensation of gate pulse periods produces exact output voltage and does not need additional circuits. The propose algorithm is verified through the simulation and experiments.

• 한국전산유체공학회:학술대회논문집
• /
• 2010.05a
• /
• pp.167-173
• /
• 2010

In this study, flow structure in a three-stage centrifugal compressor for LNG Plant with the refrigerant, Propane, was numerically investigated at the design point of the compressor using a commercial code. Flow characteristics in the passages of impeller and vaneless diffuser were analyzed in detail including velocity vector, Mach number and pressure contours in blade spanwise and meridional plane for each stage. The estimation on the one-dimensional output from the preliminary design and three-dimensional shape of the impeller blade was performed through the flow analysis. The verification for designed compressor was carried out from three-dimensional Navier-Stokes analysis. The results will be used as reference data for a new design of 3-D impeller shape to improve propane refrigerant compressor performance.

• Journal of Sensor Science and Technology
• /
• v.6 no.2
• /
• pp.106-114
• /
• 1997

In this paper, we propose an efficient coding method of remotely sensed satellite image using region classification and interband prediction. This method classifies each pixel vector considering spectral characteristics of satellite image data. Then we perform the classified intraband VQ to remove spatial (intraband) redundancy for a reference band image. To remove interband redundancy effectively, we perform the classified interband prediction for the remaining band images. Experiments on LANDSAT TM satellite image show that coding efficiency of the proposed method is better than that of the conventional Gupta's method.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.9 no.2
• /
• pp.81-86
• /
• 2014

In the scalable video coding standard, inter-layer prediction based on the coding information of the base layer was adopted to increase the coding performance. This prediction tool results in new syntax elements called motion_prediction_flag (mPF) and residul_prediction_flag(rPF), which are carried to notify the motion vector predictor (MVP) and reference block required in the motion compensation of the decoder. In this paper, an efficient coding method for mPF is proposed to enhance coding efficiency of the salable video coding standard. Through an analysis on the transmission of mPF based on the relationship between the MVPs, we discover the conditions where mPF is unnecessary at the decoder and suggest a modified rate-distortion (RD) cost function to make RD optimization more effective. Simulation results show that the proposed method offers BD rate savings of approximately 1.4%, compared with the conventional SVC standard.

• Proceedings of the KIPE Conference
• /
• 2003.07b
• /
• pp.560-564
• /
• 2003

This paper proposes a simple control strategy based on the discontinuous PWM(DPWM) to balance the DC-link voltage of three-level Neutral-Point-Clamped(WPC) inverters at low modulation index. New DPWM methods in multi-level inverter are also introduced. The proposed DPWM method changes the path and duration to flow the neutral point current out of or into neutral point of the DC-link and it makes the overall fluctuation of the DC-link voltage zero during a sampling time of reference voltage vector. Therefore, the voltage of the DC-link can be balanced fairly well and also the voltage ripple of the DC-link is reduced significantly. Moreover, comparing with conventional methods, the proposed strategy is very simple. The validity of the proposed DPWM method is verified by experiment

• The Transactions of the Korean Institute of Power Electronics
• /
• v.5 no.2
• /
• pp.115-122
• /
• 2000

In design of a digital current controller for a 3cP PWM converter, its conventional model is used in obtaining i its discretized version. But, it has errors since the characteristics of SVPWM and time delay are not taken i into consideration. In this paper, the new reference frame model of a 3c~ PWJVI converter is proposed C considering these problems. Also, the direct digital current controller based on this model is designed without a any extra algolithm. A simple tuning algorithm for the proposed current controller is given to compensate the i inductaηce mismatch problem. Then simulation and experimental results show the validity of the algolithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39A no.4
• /
• pp.209-211
• /
• 2014

This paper proposes a fast motion estimation method which can reduce the computational complexity of HEVC encoding process. While the previous method determines its search range based on a distance between a current and a reference pictures to accelerate the time-consuming motion estimation, the proposed method adaptively sets the search range according to motion vector difference between prediction units. Experimental results show that the proposed method achieves about 10.7% of reduction in processing time of motion estimation under the random access configuration whereas its coding efficiency loss is less than 0.1%.