• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.16 no.4
• /
• pp.246-253
• /
• 2016

Infrared (IR) imaging has been researched for various applications such as surveillance. IR radiation has the capability to detect thermal characteristics of objects under low-light conditions. However, automatic segmentation for finding the object of interest would be challenging since the IR detector often provides the low spatial and contrast resolution image without color and texture information. Another hindrance is that the image can be degraded by noise and clutters. This paper proposes multi-level segmentation for extracting regions of interest (ROIs) and objects of interest (OOIs) in the IR scene. Each level of the multi-level segmentation is composed of a k-means clustering algorithm, an expectation-maximization (EM) algorithm, and a decision process. The k-means clustering initializes the parameters of the Gaussian mixture model (GMM), and the EM algorithm estimates those parameters iteratively. During the multi-level segmentation, the area extracted at one level becomes the input to the next level segmentation. Thus, the segmentation is consecutively performed narrowing the area to be processed. The foreground objects are individually extracted from the final ROI windows. In the experiments, the effectiveness of the proposed method is demonstrated using several IR images, in which human subjects are captured at a long distance. The average probability of error is shown to be lower than that obtained from other conventional methods such as Gonzalez, Otsu, k-means, and EM methods.

• Smart Structures and Systems
• /
• v.16 no.2
• /
• pp.329-345
• /
• 2015

As commonly known, sensor errors and faulty signals may potentially lead structures in vibration to catastrophic failures. This paper presents a new approach to deal with sensor errors/faults in vibration control of structures by using the Fault detection and isolation (FDI) technique. To demonstrate the effectiveness of the approach, a space truss structure with semi-active devices such as Magneto-Rheological (MR) damper is used as an example. To address the problem, a Linear Matrix Inequality (LMI) based fixed-order $H_{\infty}$ FDI filter is introduced and designed. Modeling errors are treated as uncertainties in the FDI filter design to verify the robustness of the proposed FDI filter. Furthermore, an innovative Fuzzy Fault Tolerant Controller (FFTC) has been developed for this space truss structure model to preserve the pre-specified performance in the presence of sensor errors or faults. Simulation results have demonstrated that the proposed FDI filter is capable of detecting and isolating sensor errors/faults and actuator faults e.g., accelerometers and MR dampers, and the proposed FFTC can maintain the structural vibration suppression in faulty conditions.

• Journal of Navigation and Port Research
• /
• v.29 no.6 s.102
• /
• pp.487-493
• /
• 2005

Since the destruction of World Trade Center the attention of the United States and the wider international community has focussed upon the need to strengthen security and prevent terrorism This paper suggests an analysis prior to risk factor and structure for anti-terrorism in the korean maritime society. For this, in this paper, maritime terror risk factor was extracted by type and case of terror using brainstorming method. Also, risk factor is structured by FSM method and analyzed for ranking of each risk factor by AHP. At the result, the evaluation of risk factor is especially over maximum factor for related external impact.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.5 no.1
• /
• pp.52-58
• /
• 2005

We have developed a Group Method of Data Handling (GMDH) type algorithm for designing multi-layered neural networks. The algorithm is general enough that it will accept any number of inputs and any sized training set. Each neuron of the resulting network is a function of two of the inputs to the layer. The equation for each of the neurons is a quadratic polynomial. Several forms of the equation are tested for each neuron to make sure that only the best equation of two inputs is kept. All possible combinations of two inputs to each layer are also tested. By carefully testing each resulting neuron, we have developed an algorithm to keep only the best neurons at each level. The algorithm's goal is to create as accurate a network as possible while minimizing the size of the network. Software was developed to train and simulate networks using our algorithm. Several applications were modeled using our software, and the result was that our algorithm succeeded in developing small, accurate, multi-layer networks.

• Journal of the Korean Institute of Navigation
• /
• v.25 no.4
• /
• pp.477-486
• /
• 2001

Although the ports are actually competing with various strategies, the definition and structural understanding of port competitive power are not known very much. Therefore this study has launched from this fact, and has the objective of obtaining the structural model of the competitive power, and understanding the components of the port competitive power. The following are the results of the study. First, the process began by abstracting the components that composed the port competitive power through recent research, and grouping it by the most core components using the KJ method. Also, by using the FSM(Fuzzy Structural Modeling) method to understand the structure of the grouped components, and the structural model of the port competitive power was able to obtain as the result. Second, when analyzing the obtained structural model, port expenses, main trunk location, port congestion and port facility came out to be the most important component groups, and especially port expenses was the most effective component that effected all the other components overall. Third, the component groups that were relatively less important, effected by most of the other components, and located on the top level of the structure model were the hinterland accessibility, port ownership, customs duties speed, and large ship port entrance possibility etc. Fourth, the results of this study will be able to be used when establishing competing strategies for our country's ports by proposing the relatively important components with the port competitive rower considered.

• International Journal of Ocean System Engineering
• /
• v.1 no.4
• /
• pp.192-197
• /
• 2011

This paper describes the mathematical modeling, control algorithm, system design, hardware implementation and experimental test of a Manta-type Unmanned Underwater Vehicle (MUUV). The vehicle has one thruster for longitudinal propulsion, one rudder for heading angle control and two elevators for depth control. It is equipped with a pressure sensor for measuring water depth and Doppler Velocity Log for measuring position and angle. The vehicle is controlled by an on-board PC, which runs with the Windows XP operating system. The dynamic model of 6DOF is derived including the hydrodynamic forces and moments acting on the vehicle, while the hydrodynamic coefficients related to the forces and moments are obtained from experiments or estimated numerically. We also utilized the values obtained from PMM (Planar Motion Mechanism) tests found in the previous publications for numerical simulations. Various controllers such as PID, Sliding mode, Fuzzy and $H{\infty}$ are designed for depth and heading angle control in order to compare the performance of each controller based on simulation. In addition, experimental tests are carried out in a towing tank for depth keeping and heading angle tracking.

• Proceedings of the Korea Contents Association Conference
• /
• 2005.11a
• /
• pp.572-575
• /
• 2005

In digital contents industry there is a high demand to convincingly mimic the appearance and behavior of natural phenomena such as smoke, waterfall, rain, and fire. Particle systems are methods adequate for modeling fuzzy objects of natural phenomena. It is clear which parameter of which action in a particular effect should be modified for a particular visual result. The generator is usable for offline animation and for real-time special effects in digital contents and virtual reality. The application programmer is able to specify different accuracy needs for different effects. This paper design special effect generator for make low-price and high-quality digital contents in reflection industry and virtual reality applications.

• Journal of Power Electronics
• /
• v.18 no.3
• /
• pp.766-777
• /
• 2018

In this paper, a sliding mode and proportional plus integral (SM-PI) control combined with self-sustained phase shift modulation (SSPSM) for LLC resonant converters is presented. The proposed control scheme improves the transient response while preserving good steady-state performance. An averaged large signal model of an LLC converter with the ZVS modulation technique is developed for the SM control design. The sliding surface is obtained based on the input-output linearization concept. A system identification method is adopted to obtain the transform function of the LLC resonant converter, which is used to design the PI control. In order to reduce the inherent chattering problem in the steady state, the combined SM-PI control strategy is derived with fuzzy control, where the SM control is responsive during the transient state while the PI control prevails in the steady state. The combination of SSPSM and the SM-PI control provides ZVS operation, robustness and a fast transient response against step load variations. Simulation and experimental results validate the theoretical analysis and the attractive features of the proposed scheme.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.16 no.3
• /
• pp.163-172
• /
• 2016

Electric load forecasting is essential for effective power system planning and operation. Complex and nonlinear relationships exist between the electric loads and their exogenous factors. In addition, time-series load data has non-stationary characteristics, such as trend, seasonality and anomalous day effects, making it difficult to predict the future loads. This paper proposes a locally-weighted polynomial neural network (LWPNN), which is a combination of a polynomial neural network (PNN) and locally-weighted regression (LWR) for daily shortterm peak load forecasting. Model over-fitting problems can be prevented effectively because PNN has an automatic structure identification mechanism for nonlinear system modeling. LWR applied to optimize the regression coefficients of LWPNN only uses the locally-weighted learning data points located in the neighborhood of the current query point instead of using all data points. LWPNN is very effective and suitable for predicting an electric load series with nonlinear and non-stationary characteristics. To confirm the effectiveness, the proposed LWPNN, standard PNN, support vector regression and artificial neural network are applied to a real world daily peak load dataset in Korea. The proposed LWPNN shows significantly good prediction accuracy compared to the other methods.

• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2821-2823
• /
• 2002

In this paper we propose the Multi-FNN (Fuzzy-Neural Networks) for optimal identification modeling of complex system. The proposed Multi-FNNs is based on a concept of FNNs and exploit linear inference being treated as generic inference mechanisms. In the networks learning, backpropagation(BP) algorithm of neural networks is used to updata the parameters of the network in order to control of nonlinear process with complexity and uncertainty of data, proposed model use a HCM(Hard C-Means)clustering algorithm which carry out the input-output dat a preprocessing function and Genetic Algorithm which carry out optimization of model The HCM clustering method is utilized to determine the structure of Multi-FNNs. The parameters of Multi-FNN model such as apexes of membership function, learning rates, and momentum coefficients are adjusted using genetic algorithms. An aggregate performance index with a weighting factor is proposed in order to achieve a sound balance between approximation and generalization abilities of the model. NOx emission process data of gas turbine power plant is simulated in order to confirm the efficiency and feasibility of the proposed approach in this paper.