• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.4
• /
• pp.410-415
• /
• 2006

The SVDD(support vector data description) is one of the most important one-class support vector learning methods, which depends on the strategy of utilizing the balls defined on the feature space to discriminate the normal data from all other possible abnormal objects. This paper addresses on the extension of the SVDD method toward the problem of recovering the normal contents from the data contaminated with noises. The validity of the proposed de-noising method is shown via application to recovering the high-resolution images from the low-resolution images based on the high-resolution training data.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.12 no.5
• /
• pp.694-704
• /
• 2001

In this work, We evaluate the performance of eigencanceller which can suppress directional interferences and noise effectively while maintaining specified beam pattern constraints. The constraints and optimal weight vector of eigencanceller vary by using interference and noise or desired signal, interference and noise as array input signal. From the analysis results in the steady state, We show that weight vectors in each case are simplified the form of projection equation that belongs to desired subspace orthogonal to interference subspace and eigencanceller has the better performance than DMI method through mathematical analysis and simulation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.10
• /
• pp.2072-2078
• /
• 2011

In photovoltaic(PV) system, for obtaining maximum efficiency of solar power systems, the solar tracking system must be controlled to match position of the sun. In this paper, we design the solar tracking system to track movement of the sun using CdS sensor modules and to determine direction of the sun under shadow of directions. In addition, for an intelligent computation in tracking of the sun, a fuzzy controller is allocated to space avaliable for splitting area of fuzzy part for the fuzzy input space(grid-type fuzzy partition) in which a fuzzy grid partition divides fuzzy rules bases. As well, a simple model of solar tracking system is designed by two-axis motor control systems and the 8-direction sensor module that can measure shadow from CdS sensor modules by matching of axis of CdS modules and PV panels. We demonstrate this systems is effective for fixed location and moving vessels and our fuzzy controller can track the satisfactorily.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.413-415
• /
• 2022

Since human joints can be used as useful information for analyzing human behavior as a component of the human body, many studies have been conducted on human action recognition using joint information. However, it is a very complex problem to recognize human action that changes every moment using only each independent joint information. Therefore, an additional information extraction method to be used for learning and an algorithm that considers the current state based on the past state are needed. In this paper, we propose a human action recognition technique considering the positional relationship of connected joints and the change of the position of each joint over time. Using the pre-trained joint extraction model, position information of each joint is obtained, and bone information is extracted using the difference vector between the connected joints. In addition, a simplified neural network is constructed according to the two types of inputs, and spatio-temporal features are extracted by adding LSTM. As a result of the experiment using a dataset consisting of 9 behaviors, it was confirmed that when the action recognition accuracy was measured considering the temporal and spatial relationship features of each joint, it showed superior performance compared to the result using only single joint information.

• Journal of KIISE:Software and Applications
• /
• v.37 no.5
• /
• pp.363-369
• /
• 2010

Demonstration-based learning has the advantage that a user can easily teach his/her robot new task knowledge just by demonstrating directly how to perform the task. However, many previous demonstration-based learning techniques used a kind of attribute-value vector model to represent their state spaces and policies. Due to the limitation of this model, they suffered from both low efficiency of the learning process and low reusability of the learned policy. In this paper, we present a new demonstration-based learning method, in which the relational model is adopted in place of the attribute-value model. Applying the relational instance-based learning to the training examples extracted from the records of the user demonstrations, the method derives a relational instance-based policy which can be easily utilized for other similar tasks in the same domain. A relational policy maps a context, represented as a pair of (state, goal), to a corresponding action to be executed. In this paper, we give a detail explanation of our demonstration-based relational policy learning method, and then analyze the effectiveness of our learning method through some experiments using a robot simulator.

• Journal of the Korean GEO-environmental Society
• /
• v.4 no.3
• /
• pp.79-88
• /
• 2003

This paper presents a numerical study of the spatial behavior of a linear absorbing solute in a heterogeneous porous medium. The spatially correlated log-normal hydraulic conductivity field is generated in a given two-dimensional domain by using the geostatistical method (Turning Bands algorithm). The velocity vector field is calculated by applying the two-dimensional saturated groundwater flow equation to the Galerkin finite element method. The simulation of solute transport is carried out by using the random walk particle tracking model with CD(constant displacement) scheme in which the time interval is automatically adjusted. In this study, the spatial behavior of a solute is analyzed by the longitudinal center-of-mass displacement, longitudinal spatial spread moment and longitudinal plume skewness.

• Proceedings of the KIPE Conference
• /
• 2012.11a
• /
• pp.101-102
• /
• 2012

본 논문에서는 3상 승압형 정류기를 위한 슬라이딩모드 기반의 센서리스 직접전력제어방법을 제안한다. 공간벡터변조방식을 이용한 직접전력제어기법은 히스테리시스 제어기를 사용한 방법에 비해 스위칭 주파수가 일정하게 유지되어 시스템 안정성은 향상되지만, 응답속도가 떨어지는 단점이 있다. 제안하는 방법은 센서리스 기법을 이용하여 전압 센서를 제거하였고, 슬라이딩모드 제어기를 이용하여 시스템의 응답속도를 개선하였다. 3상 승압형 정류기의 시뮬레이션 결과를 통해 제안하는 제어기법의 타당성을 검증한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.11a
• /
• pp.7-10
• /
• 2002

본 논문에서는 MPEG-7의 활동도 기술자를 이용한 비디오 기술을 제안한다. 제안한 방법은 압축상태의 비디오 자료에서 직접 움직임 벡터들을 추출, 각 프레임들의 활동도의 강도를 계산하고 프레임의 흐름에 따라 계산된 활동도의 변화량에 대해 퓨리에 변환을 적용하여 얻어진 주파수 성분을 분석하여 활동도의 시간적 분포도를 계산한다. 계산된 강도 및 분포도는 MPEG-7의 표준에 따르기 위해 양자화하여 비디오 요약에 이용한다.

• Proceedings of the KIPE Conference
• /
• 2014.07a
• /
• pp.486-487
• /
• 2014

본 논문은 멀티레벨 인버터를 이용한 무정전 전원장치 시스템 (UPS; Uninterruptible Power Supply)의 신뢰성 향상을 위한 중성점 전압 불평형 제어 기법을 제안한다. 무정전 전원장치 시스템 (UPS)은 불평형 부하조건 시 직류단에 전압 불평형이 발생되며 이는 출력 전류의 왜곡을 일으켜, 민감한 부하에 손상을 입힐 수 있다. 제안한 제어기는 커패시터 전압 측정을 통해 시간오프셋을 추정한 후 공간 벡터 변조 방식의 변형 없이 중성점 전압 불평형을 제어하여, 직류단의 전압을 평형상태로 유지한다. 또한, 100kW급 UPS 시스템 시뮬레이션 검증을 통해 제어기의 타당성을 입증하였다.

• The Journal of the Korea Contents Association
• /
• v.7 no.3
• /
• pp.9-16
• /
• 2007

This paper describe methodology that is distributed on 2-dimensional plane to much high-dimensional facial motion datas using Isomap algorithm, and user interface techniques to control facial expressions by selecting expressions while user navigates this space in real-time. Isomap algorithm is processed of three steps as follow; first define an adjacency expression of each expression data, and second, calculate manifold distance between each expressions and composing expression spaces. These facial spaces are created by calculating of the shortest distance(manifold distance) between two random expressions. We have taken a Floyd algorithm for it. Third, materialize multi-dimensional expression spaces using Multidimensional Scaling, and project two dimensions plane. The smallest adjacency distance to define adjacency expressions uses Pearson Correlation Coefficient. Users can control facial expressions of 3-dimensional avatar by using user interface while they navigates two dimension spaces by real-time.