• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.1857-1858
• /
• 2008

본 논문에서는 FCM 기반 RBF 뉴럴네트워크(FCM-RBFNN) 구조를 제안하고 PSO를 이용한 FCM-RBFNN의 구조 및 파라미터의 최적화 방법을 제시한다. 클러스터링 알고리즘은 퍼지 뉴럴 네트워크에서 멤버쉽함수의 중심점과 반경 등을 결정하는 학습에 일반적으로 사용된다. 제안된 FCM-RBFNN서는 방사기저함수로써 가우시안, 삼각형 타입 등의 정해진 형태를 사용하지 않고 데이터들 사이의 거리에 관계된 계산을 수행하는 FCM에 의해 결정된다. 기존의 RBFNN에서 후반부는 상수형태로써 방사기저함수의 선형결합으로써 표현되는 반면에 제안된 FCM-RBFNN의 후반부는 상수형, 선형, 2차식 등의 다양한 형태의 다항식으로 표현될 수 있으며 다항식의 계수는 WLSE를 이용하여 추정한다. FCM 기반 RBF 뉴럴 네트워크의 성능은 퍼지규칙의 수, 후반부 다항식의 차수 FCM의 퍼지화 계수에 의하여 결정기 때문에 FCM-RBFNN의 구조와 파라미터의 최적화가 요구된다. 본 논문에서는 PSO를 이용하여 FCM-RBFNN의 구조에 관련된 퍼지 규칙의 수, 후반부 다항식의 차수와 파라미터에 관련된 퍼지화 계수를 최적화한다. 또한 후반부 다항식의 계수는 WLSE를 사용하여 추정한다.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.5
• /
• pp.19-23
• /
• 2013

TIn this paper, we propose a novel fuzzy stretching method that adopts a triangle type fuzzy membership function to control the highest and lowest brightness limit dynamically. As an essential procedure to enhance the brightness contrast, stretching is an important procedure in color image processing. While popular Ends-in Search stretching method should be provided fixed minimum and maximum brightness threshold from experience, our proposed method determines them dynamically by fuzzy membership functions. The minimum and maximum limit is determined by computing the lowest and highest pixel value according to the membership degree of our designed triangle type membership function. The experiment shows that the proposed method result in far less skewed histogram than those of Ends-in Search stretching thus successfully verifies its effectiveness.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.15 no.3
• /
• pp.119-136
• /
• 2012

The goal of this study is to apply the IPCC(Intergovernmental Panel on Climate Change) concept of vulnerability to climate change and verify the use of a combination of vulnerability index and fuzzy logic to flood vulnerability analysis and mapping in Seoul using GIS. In order to achieve this goal, this study identified indicators influencing floods based on literature review. We include indicators of exposure to climate(daily max rainfall, days of 80mm over), sensitivity(slope, geological, average DEM, impermeability layer, topography and drainage), and adaptive capacity(retarding basin and green-infra). Also, this research used fuzzy model for aggregating indicators, and utilized frequency ratio to decide fuzzy membership values. Results show that the number of days of precipitation above 80mm, the distance from river and impervious surface have comparatively strong influence on flood damage. Furthermore, when precipitation is over 269mm, areas with scare flood mitigation capacities, industrial land use, elevation of 16~20m, within 50m distance from rivers are quite vulnerable to floods. Yeongdeungpo-gu, Yongsan-gu, Mapo-gu include comparatively large vulnerable areas. This study improved previous flood vulnerability assessment methodology by adopting fuzzy model. Also, vulnerability map provides meaningful information for decision makers regarding priority areas for implementing flood mitigation policies.

• Journal of Navigation and Port Research
• /
• v.42 no.6
• /
• pp.421-426
• /
• 2018

With regard to LED lighting devices which have currently been commercialized, LED operating sequences are being sold in a fixed state. In such a state, the external environmental factors are not taken into consideration as only the illumination environment application is considered. Currently, it is difficult to create an optimal lighting environment which can adapt to changes in external environmental factors in the ship. Therefore, it was concluded that there is a need to input the external environment value so that the optimal illumination value can be reflected in real time in order to adapt more organically and actively to the change of external environmental factors. In this paper, we used a microprocessor as an integrated management system for environmental data that changes in real time according to existing external environmental factors. In addition, a controller capable of lighting control of RGB LED module by combining fuzzy inference system. For this, a fuzzy control algorithm is designed and a fuzzy control system is constructed. The distance and the illuminance value from the external environment element are input to the sensor, and these values are converted to the optimum illumination value through the fuzzy control algorithm, and are expressed through the dimming control of the RGB LED module and the practical effectiveness of the fuzzy control system is confirmed.

• Journal of Korean Society of Transportation
• /
• v.20 no.1
• /
• pp.65-76
• /
• 2002

This study is realized using fuzzy evaluation and AHP(the analytic hierarchy process) for the optimum search of traffic route and estimated by the quantitative analysis in the vague subjective judgement. It is different from classical route search and noticed thinking method of human. Appraisal element, weight, appraisal value of route is extracted from basic of the opinion gathering fur the driving expert and example of route model was used for the finding of practice utility. Model assessment was performed attribute membership function making of estimate element, estimate value setting, weight define by the AHP, non additive presentation of weight according to $\lambda$-fuzzy measure, Choquet fuzzy integral.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.1 s.33
• /
• pp.133-140
• /
• 2005

If a Player car precedes from start to the end or a fuzzy car wins every day during computer or racing games, most players will lose their interest in the games soon after several times. In order to solve this Problem and increase amusement at racing games, the more important thing than anything else is decide the ranking. In this thesis, in order to give amusement In racing games, the researcher made a fuzzy car and made it race with player cars. Because the preceding fuzzy car runs ahead of player cars, it can recognize their behaviors according to change of following player cars' speed and distance, and the fuzzy car changes its memory, but doesn't enforce actual behaviors. If the fuzzy car would make decision, it has to do behaviors to compete the ranking on the basis of the contracts it has memorized under the situation where a timer is awarded. In addition, although an accompanying fuzzy car has different contents of memory, it is operated in the same way as mentioned above. At the time of experiments, the researcher applied the actual value to the test program and drew result for ranking competition. In conclusion, the researcher could confirm that we can have modeling of various behaviors by means of the method using fuzzy logic rather than simple if-then method.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.6
• /
• pp.661-666
• /
• 2003

In this paper, the conventional k-modes and fuzzy k-modes algorithms for clustering categorical data is extended by representing the clusters of categorical data with fuzzy centroids instead of the hard-type centroids used in the original algorithm. The hard-type centroids of the traditional algorithms had difficulties in dealing with ambiguous boundary data, which might be misclassified and lead to thelocal optima. Use of fuzzy centroids makes it possible to fully exploit the power of fuzzy sets in representing the uncertainty in the classification of categorical data. The distance measure between data and fuzzy centroids is more precise and effective than those of the k-modes and fuzzy k-modes. To test the proposed approach, the proposed algorithm and two conventional algorithms were used to cluster three categorical data sets. The proposed method was found to give markedly better clustering results.

• 전자공학회논문지 IE
• /
• v.43 no.4
• /
• pp.128-135
• /
• 2006

This paper proposes the control method of fuzzy based sensor fusion by using the self localization of environment, position data by dead reckoning of the encoder and world map from sonic sensors. The proposed fuzzy based sensor fusion system recognizes the object and extracts features such as edge, distance and patterns for generating the world map and self localization. Therefore, this paper has developed fuzzy based control of mobile robot with experimentations in a corridor environment.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.11a
• /
• pp.320-323
• /
• 2007

많은 센서 네트워크 응용분야에서 센서 노드는 개방된 환경에 놓이게 된다. 공격자는 개방된 환경에 놓인 센서 노드를 물리적으로 획득할 수 있으며, 포획한 노드를 이용하여 허위보고서를 센서네트워크에 삽입 시킬 수 있다. 삽입된 허위 보고서는 제한된 센서노드 에너지를 고갈 시키며, 허위 경보를 일으켜 심각한 문제를 야기 시킬 수도 있다. 이러한 공격을 막기 위해 Yu와 Guan은 Dynamic En-route Filtering(DEF) 방법을 제안 하였다. DEF는 인증키 재분배를 통하여 필터링 효율을 강화 시키지만, 키 분배 효율을 높일수록 인증키를 분배에 사용되는 에너지 소비가 커지는 문제점을 가지고 있다. 본 논문에서는 인증키 분배에 소비되는 에너지를 줄이면서 효율적인 키 분배를 위해 퍼지 시스템 기반의 키 분배 거리 결정 방법을 제안하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.371-373
• /
• 1999

축구경기는 세계적으로 널리 알려진 스포츠이다. 특히, 한국과 일본은 2002년 월드컵을 개최하는 나라이다. 이와 같은 배경에서, 우리는 퍼지 논리를 이용하여 축구 선수들의 기본적인 동작(Motion)을 인식하는 알고리즘을 제안한다. 여기서 기본 동작이란, 걷기, 뛰기, 드리블하기, 서있기 등으로 정의한다. 만약 우리가 축구경기 장면 비디오를 통해 동작 인식을 필요로 한다면, 특징 패턴은 축구선수들의 양다리 각과 한 축구선수와 공과의 거리가 중요 요소가 될 것이다. 이 논문은 축구 선수 및 공의 추적(Tracking)을 통해 전처리된 데이터를 얻을 수 있다고 가정한다. 실험결과를 통해 축구선수 뿐만 아니라 스포츠 경기에서 선수들의 기본 동작 인식에 퍼지 논리를 적용할 수 있음을 알 수 있었다.