• Proceedings of the Korean Society of Computer Information Conference
• /
• 2009.01a
• /
• pp.25-30
• /
• 2009

패턴 인식에서 분류기 모형으로 많이 사용되는 퍼지 가중치 평균 분류기는 가중치를 적절히 설정함으로써 뛰어난 분류 성능을 얻을 수 있다는 장점이 있다. 그러나 일반적으로 가중치는 인식 문제 분야의 특성이나 해당 전문가의 지식이나 주관적 경험을 기반으로 설정되므로 설정된 가중치의 일관성과 객관성을 보장하기가 어려운 문제점을 갖고 있다. 따라서 이 논문에서는 퍼지 가중치 평균 분류기의 가중치를 설정하기 위한 객관적 기준을 제시하기 위하여 특징값들 간의 통계적 정보를 이용한 가중치 설정 기법들을 제안하였다. 제안한 기법들을 이용하여 UCI machine learning repository 사이트에서 제공되는 표준 데이터들 중의 하나인 Iris 데이터 세트를 이용하여 실험하고 그 결과를 비교, 분석하였다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.2
• /
• pp.161-165
• /
• 2014

In this paper, we proposed the fuzzy prototype pattern classifier. In the proposed classifier, each prototype is defined to describe the related sub-space and the weight value is assigned to the prototype. The weight value assigned to the prototype leads to the change of the boundary surface. In order to define the prototypes, we use Fuzzy C-Means Clustering which is the one of fuzzy clustering methods. In order to optimize the weight values assigned to the prototypes, we use the Differential Evolutionary Algorithm. We use Linear Discriminant Analysis to estimate the coefficients of the polynomial which is the structure of the consequent part of a fuzzy rule. Finally, in order to evaluate the classification ability of the proposed pattern classifier, the machine learning data sets are used.

• Journal of KIISE:Software and Applications
• /
• v.33 no.5
• /
• pp.489-498
• /
• 2006

In this paper, we suggested a vehicle classification algorithm using pattern recognition method. At present, Inductive Loop Detector is rarely used for vehicle classification because of its low accuracy. To improve the accuracy, we suggest a new algorithm for Loop Detector using neural networks. In the developed algorithm, the inputs to the neural networks are the variation rate of frequency and occupancy-time. The output is classified vehicles. The developed algorithm was assessed at test sites and the recognition rate was 91.3percent. The results verified that the proposed algorithm improves the vehicle classification accuracy compared to the conventional method based on Loop Detector.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.5
• /
• pp.931-936
• /
• 2009

Classification problem plays a major role in grouping of the records in database systems, detection of faults in VLSI circuits, image processing, and so on. In this paper, we propose the algorithm constructing D1-MACA as a two-class classifier with minimum memory for given pattern sets using the concepts of subspace. Also we analyze the condition that is designed a two-class classifier D1-MACA with two attractors.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.05a
• /
• pp.123-125
• /
• 2002

여러 사람에게서 생체신호를 측정하여 특징을 추출하는 경우 피실험자마다 다른 신체적 또는 생리학적 특징에 의해 같은 클래스로 분류하고 싶어도 다른 클래스로 잘못 분류되는 경우가 발생한다. 이와 같이 N 명의 사람에게서 얻은 생체신호로 M 개의 클래스를 분류하도록 훈련하여 새로운 사람의 생체신호를 M 개의 클래스로 분류하고자 할 때 발생하는 문제를 해결하기 위한 방법으로 피실험자 독립적인 클러스터링 방법을 제안하고자 한다. 이를 위한 수학적 기반으로 동치관계들의 교집합과 합집합에 근거한 새로운 연산자를 정의하고 이를 이용하여 최대 공통 클러스터(Largest Common Cluster, LCC)라는 새로운 개념을 정의한다 이는 여러 사람에게서 얻은 정보에서 최대한 공통의 성질을 갖는 것들을 찾아내는 수학적이고 체계적인 방법이라 할 수 있다. 따라서 일단 LCC를 찾아내면 이를 특징(feature)으로 삼아 패턴분류기를 설계하면 여러 사람에게 적용가능한 생체신호 인식기를 설계할 수 있게 된다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.11a
• /
• pp.279-282
• /
• 2007

Support Vector Machine(SVM)은 통계적 학습 이론에 기반을 둔 분류기이다. 또한 Twin Support Vector Machine(TWSVM)은 이진 SVM 분류기의 한 종류로써, 서로 관련된 두 개의 SVM 유형 문제를 통해 평행하지 않은 두 개의 평면을 결정하고 이 두 평면을 통해 분류기를 완성하는 방식이다. 이러한 방식은 TWSVM은 학습 시간이 SVM에 비해 훨씬 짧으며, SVM과 비교하여 떨어지지 않는 성능을 보여준다. 본 논문은 분류기 입력에 Fuzzy Memvership을 적용하는 방식의 TWSVM을 제안하고, 2차원 벡터 입력에 대한 실험을 통하여 기존에 제시 되었던 TWSVM과 비교한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.2
• /
• pp.120-126
• /
• 2016

In this study, we propose a method for effectively detecting and recognizing the face in image using RBFNNs pattern classifier and HCbCr-based skin color feature. Skin color detection is computationally rapid and is robust to pattern variation for face detection, however, the objects with similar colors can be mistakenly detected as face. Thus, in order to enhance the accuracy of the skin detection, we take into consideration the combination of the H and CbCr components jointly obtained from both HSI and YCbCr color space. Then, the exact location of the face is found from the candidate region of skin color by detecting the eyes through the Haar-like feature. Finally, the face recognition is performed by using the proposed FCM-based RBFNNs pattern classifier. We show the results as well as computer simulation experiments carried out by using the image database of Cambridge ICPR.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.4
• /
• pp.355-360
• /
• 2015

In this paper, we propose the design of Radial Basis Function Neural Network based on PCA in order to recognize handwritten digits. The proposed pattern classifier consists of the preprocessing step of PCA and the pattern classification step of pRBFNNs. In the preprocessing step, Feature data is obtained through preprocessing step of PCA for minimizing the information loss of given data and then this data is used as input data to pRBFNNs. The hidden layer of the proposed classifier is built up by Fuzzy C-Means(FCM) clustering algorithm and the connection weights are defined as linear polynomial function. In the output layer, polynomial parameters are obtained by using Least Square Estimation (LSE). MNIST database known as one of the benchmark handwritten dataset is applied for the performance evaluation of the proposed classifier. The experimental results of the proposed system are compared with other existing classifiers.

• Journal of The Geomorphological Association of Korea
• /
• v.19 no.4
• /
• pp.139-155
• /
• 2012

Most of previous landform classification methods using DEM compares the values between the center of the cell and the surrounding cells, which in turn, greatly depends on analysis scale. To overcome the problem of scale-dependency, a new classification scheme is developed, which is called "Geomorphons". Unlike the traditional approaches using DEM, Geomorphons is the way which compares the level with other cells against the criteria cell. As a pilot study, we classify the landforms of Pyeongchang-Gun in Korea. Then, we compare the result with the other methods such as Topographic Position Index. Through the systematic analysis, we obtain the following findings. First, Geomorphons can reduce the time for the classification of landforms because of using unsupervised classification. Second, Geomorphons is little dependent on change in the scale, which can provide a pilot tool for reconnaissance study for covering large area.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.6
• /
• pp.594-600
• /
• 2017

The new design methodology of prediction model and pattern classification, which is based on the dimension reduction algorithm called principal component analysis, is introduced in this paper. Principal component analysis is one of dimension reduction techniques which are used to reduce the dimension of the input space and extract some good features from the original input variables. The extracted input variables are applied to the prediction model and pattern classifier as the input variables. The introduced prediction model and pattern classifier are based on the very simple regression which is the key point of the paper. The structural simplicity of the prediction model and pattern classifier leads to reducing the over-fitting problem. In order to validate the proposed prediction model and pattern classifier, several machine learning data sets are used.