• Journal of Electrical Engineering and Technology
• /
• 제12권6호
• /
• 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.

• 대한기계학회논문집A
• /
• 제27권2호
• /
• pp.310-316
• /
• 2003

One of the significant subject in the automatic arc welding is to establish control system of the welding parameters for controlling bead geometry as a criterion to evaluate the quality of arc welding. This paper proposes an inference algorithm for bead geometry in CMA Welding using Neuro-Fuzzy algorithm. The characteristic welding parameters are measured by the circuit composed of hall sensor, voltage divider tachometer, etc. and then the bead geometry of each weld pool is calculated and detected by an image processing with CCD camera and a measuring with microscope. The relationships between the characteristic welding parameters and the bead geometry have been arranged empirically. From the result of experiments, membership functions and fuzzy rules are tuned and determined by the learning of neural network, and then the relationship between actual bead geometry and inferred bead geometry are concluded by fuzzy logic controller. In the applied inference system of bead geometry using Neuro-Fuzzy algorithm, the inference error percent is within -5%∼+4% in case of bead width, -10%∼+10% in bead height, -5%∼+6% in bead area, -10%∼+10% in penetration. Use of the Neuro-Fuzzy algorithm allows the CMA Welding system to evaluate the quality in bead geometry in real time as the welding parameters change.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제2권4호
• /
• pp.239-250
• /
• 2013

가용 센서를 내장하고 있는 모바일 디바이스의 사용이 많아지고 자동화, 자율화, 사용자 맞춤식 서비스의 요구가 커짐에 따라 컨텍스트 인지 모바일 컴퓨팅 (Context-Aware Mobile Computing)의 필요성이 증대하고 있다. 하지만 추론 시스템 설계는 컨텍스트 분석, 인지하고자 하는 상황분석 등의 복잡한 과정을 요구한다. 또한 모바일 디바이스의 제한된 자원 때문에 컴퓨팅 파워가 높은 서버에 탑재된 추론 엔진을 통해 추론을 수행하는 것이 적합하다. 본 논문에서는 이러한 과정을 간결하고 정확하게 표현하기 위한 컨텍스트-상황 추론 요소의 범용적 정형 모델을 제안하고 추론 요소들의 정형 모델을 실사례에 적용하여 본 논문에서 제안하고 있는 추론 요소들의 정형 모델이 실효성을 가지고 있으며 범용적임을 보여준다. 또한 제한한 추론 요소들을 컴퓨팅 환경에서 실현화하기 위해 제안한 정형 모델들을 기반으로 추론 엔진을 설계 및 구현하고 추론 실험을 통해 추론 엔진의 실효성과 재사용성을 검증한다.

• 한국융합학회논문지
• /
• 제5권1호
• /
• pp.35-41
• /
• 2014

An automatic speech recognition system is one of the popular research problems. There are many research groups working in this field for different language including Japanese. Japanese vowel recognition is one of important parts in the Japanese speech recognition system. The vowel classification system with the Mamdani fuzzy inference system was developed in this research. We tested our system on the blind test data set collected from one male native Japanese speaker and four male non-native Japanese speakers. All subjects in the blind test data set were not the same subjects in the training data set. We found out that the classification rate from the training data set is 95.0 %. In the speaker-independent experiments, the classification rate from the native speaker is around 70.0 %, whereas that from the non-native speakers is around 80.5 %.

• 한국군사과학기술학회지
• /
• 제3권1호
• /
• pp.76-84
• /
• 2000

As first step of any acoustic defence system, a attacking target warning system needs to be extremely reliable. This means the system must ensure a high probability of target classification together with a very low false alarm rate. In this paper, a algorithms for underwater target automatic classification is available for use in the passive sonar will be presented. In first, we will describe the precise automatic extraction of frequency lines for the detection of acoustic signatures. Also, a neural network and fuzzy based algorithms for target classification will be described. Thus the performances of these algorithms are very good with a high probability of classification.

• 한국항행학회논문지
• /
• 제11권3호
• /
• pp.343-349
• /
• 2007

본 연구에서는 전형적인 패턴인식 기법을 적용한 지적도면 인식 방법의 시간지연 문제를 해결하기 위하여 퍼지추론을 이용한 지적도면 인식 방법을 제안하였다. 퍼지 입력 파라미터는 지적도면에 있는 선분의 굵기와 색, 문자 및 숫자를 활용하였다. 퍼지 관계맵(Fuzzy Association Map: FAM)을 생성하였고 추론결과 지적도에서 서비스에 필요한 정보들을 추출 할 수 있었다. 결과물은 지적도를 이용하여 건축물이 들어설 수 있는 공간을 예측하고 이를 3차원으로 자동 형성시키는 방안의 전 단계 과정인 바, u-Gov 기반의 토지 등기 열람 서비스 사업과 인터넷 민원서비스 고도화 사업과 연계하여 적용 시킬 수 있다.

• 대한교통학회지
• /
• 제14권2호
• /
• pp.119-136
• /
• 1996

The evaluation of risk level or possibility of traffic accidents is a fundamental task in reducing the dangers associated with current transportation system. However, due to the lack of data and basic researches for identifying such factors, evaluations so far have been undertaken by only the experts who can use their judgements well in this regard. Here comes the motivation this thesis to evaluate such risk level more or less in an automatic manner. The purpose of this thesis is to test the fuzzy-logic theory in evaluating the risk level of traffic accidents. In modeling the process of expert's logical inference of risk level determination, only the geometric features have been considered for the simplicity of the modeling. They are the visibility of road surface, horizontal alignment, vertical grade, diverging point, and the location of pedestrain crossing. At the same time, among some inference methods, fuzzy composition inference method has been employed as a back-bone inference mechanism. In calibration, the proposed model used four sites' data. After that, using calibrated model, six sites' risk levels have been identified. The results of the six sites' outcomes were quite similar to those of real world other than some errors caused by the enforcement of the model's output. But it seems that this kind of errors can be overcome in the future if some other factors such as driver characteristics, traffic environment, and traffic control conditions have been considered. Futhermore, the application of site's specific time series data would produce better results.

• 생물환경조절학회지
• /
• 제20권3호
• /
• pp.169-175
• /
• 2011

본 연구에서는 유비쿼터스 식물공장의 재배환경에 필요한 요소들의 센서 네트워크를 구성하고 자동으로 감지하여 적응형 뉴로-퍼지 추론시스템을 통하여 환경변화를 추론하여 식물공장의 재배환경을 적절하게 제어할 수 있는 새로운 자동제어시스템의 프레임워크를 제안하고, 이를 설계하였다. 유비쿼터스 식물공장 환경을 제어하기 위하여 식물공장의 재배환경에 영향을 미치는 환경요소인 실내온도, 근권온도, 습도, 광도, $CO_2$ 농도를 측정할 수 있는 센서 네트워크를 구성하고 측정된 환경요소의 변화에 따라 램프, 환기, 습도, $CO_2$ 농도, 온도를 제어할 수 있는 장치를 자동으로 제어할 수 있는 식물공장 자동제어시스템을 설계하였다. 이를 위하여 본 연구에서는 센서를 통하여 받아들이는 입력값을 퍼지소속함수로 변화하고 적응형 뉴로-퍼지시스템에 따라 추론하고 평가하여 보다 정밀하게 식물공장을 자동으로 제어할 수 알고리즘을 개발하였고 이를 구현하였다. 개발된 자동제어시스템을 상추 식물공장에 적용한 결과 만족스러운 시험결과를 얻을 수 있었다. 향후 연구로는 식물공장에서 재배하고 있는 작물별 생장모델의 적합도 검정 및 개선을 위하여, 작물별 재배규칙을 보다 상세히 도출하는 것이 필요하고, 작물의 재배에 필요한 지식을 보다 정량적으로 표현하고 지식상에 내포하고 있는 불확실성을 해결하는 것이 필요하다. 더 나아가 식물공장에서 환경인자간의 상호관련성을 보다 정밀하게 수식화하고 이를 추론할 수 있는 정밀하고 과학적인 자동제어시스템의 개발이 필요하다.

• International Journal of Control, Automation, and Systems
• /
• 제2권1호
• /
• pp.39-54
• /
• 2004

We developed an automatic blood pressure control system to maintain the blood pressure of patients at a substantially low level during a surgical operation. The developed system discharges two functions, continuous feedback control of the mean arterial pressure (MAP) by a state-predictive servo controller and risk control based on the inference by fuzzy-like logics and rules using measured data. Twenty-eight clinical applications were made beginning in November 1995, and the effects of the automatic blood pressure control on the operation time and on bleeding were assessed affirmatively by means of Wilcoxon testing. This paper essentially reports the engineering details of the control system.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1995년도 추계학술대회 95 KFIS Workshop Realization of Human Friendly System Based on Soft Computiong Techniques
• /
• pp.124-140
• /
• 1995

The merit of fuzzy rule based systems stems from their capability of encoding qualitative knowledge of experts into quantitative rules. Recent advancement in automatic tuning or self-organization of fuzzy rules from experimental data further enhances their power, allowing the integration of the top-down encoding of knowledge with the bottom-up learning of rules. In this paper, methods of self-organizing fuzzy rules and of performing defuzzification and inference is presented based on a multi-resolution radial basis function network. The network learns an arbitrary input-output mapping from sample distribution as the union of hyper-ellipsoidal clusters of various locations, sizes and shapes. The hyper-ellipsoidal clusters, representing fuzzy rules, are self-organized based of global competition in such a way as to ensute uniform mapping errors. The cooperative interpolation among the multiple clusters associated with a mapping allows the network to perform a bidirectional many-to-many mapping, representing a particular from of defuzzification. Finally, an inference engine is constructed for the network to search for an optimal chain of rules or situation transitions under the constraint of transition feasibilities imposed by the learned mapping. Applications of the proposed network to skill acquisition are shown.