Go to the main menu
Skip to content
Go to bottom
REFERENCE LINKING PLATFORM OF KOREA S&T JOURNALS
> Journal Vol & Issue
Journal of Korean Institute of Intelligent Systems
Journal Basic Information
Journal DOI :
Korean Institute of Intelligent Systems
Editor in Chief :
Volume & Issues
Volume 7, Issue 5 - Dec 1997
Volume 7, Issue 4 - Oct 1997
Volume 7, Issue 3 - Aug 1997
Volume 7, Issue 2 - Jun 1997
Volume 7, Issue 1 - Mar 1997
Selecting the target year
An Intellingnet Query Processing System for Relational Database System
Journal of Korean Institute of Intelligent Systems, volume 7, issue 4, 1997, Pages 1~8
In this paper, we propose a new intelligent query processing system for relational database !systems. By analyzing previous research results related with fuzzy queries, a new intelligent query processing sysytem is developed and the role of each module including intelligent query processor is defined and :some algorithms for parser, query translation module, inference engine, semantic DB and result com-poser are suggested. By applying a typical example to the proposed intelligent query processing liysytem, reasonable results for the ambiguous query are drawn, and therefore it shows a promising model returning ordered result for both the ambiguous queries and general queries.
Fuzzy Decision based on Motion Characteristics
Journal of Korean Institute of Intelligent Systems, volume 7, issue 4, 1997, Pages 9~17
This paper describes a monitoring system that examines water quality by analyzing behavioral patterns of fishes. The water quality inspection system (WQIS) captures color images of fishes with a CCD camera, extracts out fish regions from the images, and determines motion characteristics of fishes by computing consecutive frames. We define five types of measures that reflect behavioral patterns of fishes : floatness, fledness, clustemess, diffusiveness, and mobility. These measures are utilized when the system performs fuzzy inference to induce the conclusion about water quality. We believe that the proposed system can be a solution for securing clean water.
Image Compression using an Intelligne Classified Vector Quantization Method in Transform Domain
Journal of Korean Institute of Intelligent Systems, volume 7, issue 4, 1997, Pages 18~28
This paper presents image data compression using a classified vector quantization (CVQ) which categories edge blocks according to the energy distribution of subimages in the discrete cosine transform domain. Classifying the edge blocks enhances visual quality of the compressed images while maintaining a high compression ratio. The proposed classification method categories subimages into eight lypes of edge features according to an energy distribution. A neural network, trained with the data generated from the proposed classification method, can successfully classify subimages to eight edge categories. Experimental results are given to show how the (1VQ method incorporatd with a neural network can produce faithful compressed image quality for high compression ratios.
A Disctete Model Reference Control With a Neural Network System Ldentification for an Active Four Wheel Steering System
Journal of Korean Institute of Intelligent Systems, volume 7, issue 4, 1997, Pages 29~39
A discrete model reference control scheme for a vehicle four wheel steering system(4WS) is proposed and evaluated for a class of discrete time nonlinar dynamics. The schmen employs a neural network to identify the plan systems, wher the neural network estimates the nonlinear dynamics of the plant. The algorithm is proven to be globally stable, with tracking errors converging to the neighborhood of zero. The merits of this scheme is that the global system stability is guaranteed. Whith thd resulting identification model which contains the neural networks, the parameters of controller are adjusted. The proposed scheme is applied to the vehicle active four wheel system and shows the validity and effectiveness through simulation. The three-degree-of freedom vehicle handling model is used to investigate vehicle handing performances. In simulation of the J-turn maneuver, the yaw rate overshoot reduction of a typical mid-size car is improved by 30% compared to a two wheel steering system(2WS) case, resulting that the proposed scheme gives faster yaw rate response andl smaller side slip angle than the 2WS case.
Position Recognition and Leaning Correction of DNA Ban Images
Journal of Korean Institute of Intelligent Systems, volume 7, issue 4, 1997, Pages 40~47
This paper proposes a method using the straight line Hough transform(SLHT) that recognizes the position of DNA band images from the scanner. The method also detects and corrects automatically the leaning angle of the image. After binarization of a gray-scale DNA band images, the SLHT detects line components involved in the image and recognizes the position of the image using the cross paints of the line components assuming the image is in retangular shape. To improve efficiency of reading many IINA band images through the scanner, this method finds and corrects the leaning angle accurately as less than -t I degree.
The Optimization of Fuzzy Logic Controllers Using Genetic Algorithm
Journal of Korean Institute of Intelligent Systems, volume 7, issue 4, 1997, Pages 48~57
This paper presents the automatic construction and parameter optimization technique for fuzzy logic controllers using genetic algorithm. In general. the design of fuzzy logic controllers has difficulties in the acq~lisition of expert's knowledge and relies to a great extent on empirical and heuristic knowledge which, in many cases, cannot be objectively justified. So, the performance of the controllers c:an be degraded in the case of plant parameter variations or unpredictable incident which a designer may have ignored, and the parameters of fuzzy logic controllers obtained by expert's control action may not be optirnal. Some of these problems can be resolved by the use of genetic algorithm. The proposed method can tune the parameters of fuzzy logic controllers including scaling factors and determine: the appropriate number of fuzzy rulcs systematically. Finally, we provides the second order dead time plant to evaluate the feasibility and generality of the proposed method. Comparison shows that the proposed method can produce fuzzy logic controllers with higher accuracy and a smaller number of fuzzy rules than manually tuned fuzzy logic controllers.
A Study on the Performance Improvement of Fuzzy Controller Using Genetic Algorithm and Evolution Programming
Journal of Korean Institute of Intelligent Systems, volume 7, issue 4, 1997, Pages 58~64
FLC(Fuzzy Logic Controller) is stronger to the disturbance than a classical controller and its overshoot of the intialized value is excellent. In case an unknown process or the mathematical modeling of a complicated system is impossible, a fit control quantity can be acquired by the Fuzzy inference. But FLC can not converge correctly to the desirable value because the FLC's output value by the size of the quantization level of the Fuzzy variable always has a minor error. There are many ways to eliminate the minor error, but I will suggest GA-FLC and EP-FLC Hybrid controller which csombines FLC with GA(Genetic Algorithm) and EP(Evo1ution Programming). In this paper, the output characteristics of this Hybrid controller will be compared and analyzed with those of FLC, it will he showed that this Hybrid controller converge correctly to the desirable value without any error, and !he convergence speed performance of these two kinds of Hyhrid controller also will be compared.
Structure Pruning of Dynamic Recurrent Neural Networks Based on Evolutionary Computations
Journal of Korean Institute of Intelligent Systems, volume 7, issue 4, 1997, Pages 65~73
This paper proposes a new method of the structure pruning of dynamic recurrent neural networks (DRNN) using evolutionary computations. In general, evolutionary computations are population-based search methods, therefore it is very useful when several different properties of neural networks need to be optimized. In order to prune the structure of the DRNN in this paper, we used the evolutionary programming that searches the structure and weight of the DRNN and evolution strategies which train the weight of neuron and pruned the net structure. An addition or elimination of the hidden-layer's node of the DRNN is decided by mutation probability. Its strategy is as follows, the node which has mhnimum sum of input weights is eliminated and a node is added by predesignated probability function. In this case, the weight is connected to the other nodes according to the probability in all cases which can in- 11:ract to the other nodes. The proposed pruning scheme is exemplified on the stabilization and position control of the inverted-pendulum system and visual servoing of a robot manipulator and the effc: ctiveness of the proposed method is demonstrated by numerical simulations.
Evolutionary Learning Algorithm fo r Projection Neural NEtworks
Journal of Korean Institute of Intelligent Systems, volume 7, issue 4, 1997, Pages 74~81
This paper proposes an evolutionary learning algorithm to discipline the projection neural nctworks (PNNs) with special type of hidden nodes which can activate radial basis functions as well as sigmoid functions. The proposed algorithm not only trains the parameters and the connection weights hut also c~ptimizes the network structure. Through the structure optimization, the number of hidden node:; necessary to represent a given target function is determined and the role of each hidden node is decided whether it activates a radial basis function or a sigmoid function. To apply the algorithm, PNN is realized by a self-organizing genotype representation with a linked list data structure. Simulations show that the algorithm can build the PNN with less hidden nodes than thc existing learning algorithm using error hack propagation(EE3P) and network growing strategy.
Fuzzy Measure-based Subset Interactive Models for Interactive Systems.
Journal of Korean Institute of Intelligent Systems, volume 7, issue 4, 1997, Pages 82~92
In this paper, a fuzzy measure and integral-based model fnr interactive systems is proposed. The processes of model identification consists of the following three steps : (i) structure identification (ii) parameter identification and (iii) selection of an optimal model. An algorithm for the model structure identification using the well-known genetic algorithm ((;A) with a modified selection operator is proposed. A method for the identification of par;imetcrs corresponding to fuzzy measures is presented. A statistical model selection criterion is used for the selection of an optimal model among the candidates. Finally, experimental results obtained hy applying the proposed model to the subjective evaluation data set and the well-known time series data are presented to show the validity of the proposed model.