• Journal of Electrical Engineering and Technology
• /
• v.12 no.6
• /
• pp.2388-2398
• /
• 2017

In this study, we propose a robust face recognition system to pose variations based on automatic pose estimation. Radial basis function neural network is applied as one of the functional components of the overall face recognition system. The proposed system consists of preprocessing and recognition modules to provide a solution to pose variation and high-dimensional pattern recognition problems. In the preprocessing part, principal component analysis (PCA) and 2-dimensional 2-directional PCA ($(2D)^2$ PCA) are applied. These functional modules are useful in reducing dimensionality of the feature space. The proposed RBFNNs architecture consists of three functional modules such as condition, conclusion and inference phase realized in terms of fuzzy "if-then" rules. In the condition phase of fuzzy rules, the input space is partitioned with the use of fuzzy clustering realized by the Fuzzy C-Means (FCM) algorithm. In conclusion phase of rules, the connections (weights) are realized through four types of polynomials such as constant, linear, quadratic and modified quadratic. The coefficients of the RBFNNs model are obtained by fuzzy inference method constituting the inference phase of fuzzy rules. The essential design parameters (such as the number of nodes, and fuzzification coefficient) of the networks are optimized with the aid of Particle Swarm Optimization (PSO). Experimental results completed on standard face database -Honda/UCSD, Cambridge Head pose, and IC&CI databases demonstrate the effectiveness and efficiency of face recognition system compared with other studies.

• Proceedings of the KIEE Conference
• /
• 1999.11a
• /
• pp.28-30
• /
• 1999

This paper proposes a robust FPID(Fuzzy Proportional Integral Derivative) controller for the LFC(load frequency control) of 2-area power system. The PID gain parameters of the proposed robust FPID controller are self-tuned by PSGM(Product Sum Gravity Method) which is very similiar to human's inference procedures. As the results of simulation, the proposed FPID controller against various load disturbances shows that it is superior to the conventional control techniques such as optimal, PID and fuzzy control in the response characteristics of frequency and tie line power flow.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.03a
• /
• pp.35-38
• /
• 1998

Abstracts In this paper, we propose a robust adaptive fuzzy control scheme using a sliding control input for tracking of a class of MISO nonlinear systems with unknown bounded external disturbances. In the proposed scheme, the nonlinearity is estimated adaptively via a fuzzy inference based on a fuzzy model. A sliding control input is introduced such that boundedness of all signals in the system is guaranteed even though the existence of a fuzzy approximation error and external disturbances. The controller parameters are updated by using a proposed adaptation law, which is similar 1-modification method. Computer simulation shows the effectiveness of the proposed control scheme.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1999.10a
• /
• pp.42-46
• /
• 1999

Knowledge management(KM) is emerging as a robust management mechanism with which an organization can remain highly intelligent and competitive in a turbulent market. Organization memory(or knowledge) is at the heart of KM success. How to create organizational memory has been debated among researchers. In literature, a wide variety of methods for creating organizational memory have been proposed only to prove that its applicability is limited to decision-making problems which require shallow or non-causal knowledge type. However, organizational memory with a sense of causal knowledge is highly required in solving complicated decision-making problems in which complex dynamics exist between various factors and influence each other with cause and effect relationship among them. In this respect, we propose a new approach to creating a causal-typed organizational memory (CATOM), which has a form of causal knowledge and is represented in a matrix form, by using an inference diagram. An algorithm for CATOM creation is suggested and applied to an illustrative example. Results show that our proposed KM approach can effectively equip an organization with semi-automated CATOM creation and inference process which is deemed useful in a highly competitive business environment.

• ETRI Journal
• /
• v.31 no.2
• /
• pp.182-192
• /
• 2009

Rain attenuation and intercell interference are two crucial factors in the performance of broadband wireless access networks such as local multipoint distribution systems (LMDS) operating at frequencies above 20 GHz. Power control can enhance the performance of downlink CDMA-based LMDS systems by reducing intercell interference under clear sky conditions; however, it may damage system performance under rainy conditions. To ensure robust operation under both clear sky and rainy conditions, we propose a novel power-control scheme which applies an adaptive neuro-fuzzy inference system (ANFIS) for downlink CDMA-based LMDS systems. In the proposed system, the rain rate and the number of users are two inputs of the fuzzy inference system, and output is defined as channel quality, which is applied in the power control scheme to adjust the power control region. Moreover, ITU-R P.530 is employed to estimate the rain attenuation. The influence of the rain rate and the number of users on the distance-based power control (DBPC) scheme is included in the simulation model as the training database. Simulation results indicate that the proposed scheme improves the throughput of the DBPC scheme.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.207-210
• /
• 1997

In this paper, a hierarchical fuzzy controller for stabilization of the inverted pendulum system is proposed. The facility of this hierarchical fuzzy controller which has a swing-up control mode and a stabilization one, moves a pendulum in an initial natural stable equilibrium point and a cart in arbitrary position to an unstable equilibrium point and a center of rail. Specially, the virtual equilibrium point (.PHI.$_{VEq}$ ) which describes functionally considers the interactive dynamics between a position of cart and a angle of inverted pendulum is introduced. And comparing with the convention optimal controller, the proposed hierarchical fuzzy inference made substantially the inverted pendulum system robust and stable.e.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.59 no.4
• /
• pp.793-796
• /
• 2010

The equalizer compensates a signal distorted by transmission lines and amplifying stages, so the signal can have uniform characteristics over all frequency range. The equalizer in ODD(Optical Disc Drive) improves the stability of the extracted clock from a received signal and the readability of an inserted disc by suppressing noise and ISI(Inter-Symbol Inference). The length of marks-spaces and track pitch on discs becomes shorter as the recording density of an optical media is higher, which causes noise and ISI. And, the sensitivity about the fluctuation of physical systems is higher as the optical devices become more complicate. This paper proposes a method to select the coefficient of built-in equalizer of ODD in order to maintain the quality of signals against noise and ISI caused by system fluctuation.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.382-384
• /
• 2004

In this parer, we propose a clustering that perform algorithm using new convergence properties. For detection and optimization of cluster, we use to similarity measure with cumulative probability and to inference the its parameters with MLE. A merits of using the cumulative probability in our method is very effectiveness that robust to noise or unnecessary data for inference the parameters. And we adopt similarity threshold to converge the number of cluster that is enable to past convergence and delete the other influence for this learning algorithm. In the simulation, we show effectiveness of our algorithm for convergence and optimization of cluster in riven data set.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.70-73
• /
• 2003

This paper proposes a movement instruction system using virtual environment. This system consists of a monitor, cameras, ana a PC. A learner is coached by a virtual instructor that is displayed in virtual environment as 3 dimensional computer graphics on the monitor. Virtual instructor shows sample movement and suggests mistakes of learner's movement by recognizing movement of learner's movement from the picture that cameras capture. To improve the robust characteristic of information from cameras, the system enables to select optimum inputs from cameras based on learner's movement It implemented by Fuzzy associative inference system Fuzzy associative inference system is implemented by bi-directional associative memory and fuzzy rules. It is suitable to convert obscure information into clear. We implement and evaluate the movement instruction system

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.313-314
• /
• 2008

Ad hoc network is a multi-hop wireless network formed with non-infrastructure. The fact that limited resource could support the network of robust, simple framework and energy conserving etc. In this paper, we propose a new ad hoc multicast routing protocol for based on the ontology scheme called inference network. Ontology knowledge-based is one of the structure of context-aware.