• 응용통계연구
• /
• 제13권2호
• /
• pp.343-352
• /
• 2000

K-평균 군집화(K-means clustering)는 비계층적 군집화 방법이 하나로서 큰 자료에서 개체 군집화에 효율적인 것으로 알려져 있다. 그러나 종종 비교적 균일한 대군집의 일부를 소군집에 떼어주는 오류를 범하기도 한다. 이 연구에서는 그러한 현상을 정확히 인지하고 이에 대한 대책으로서 ‘이중 K-평균 군집화(double K-means clustering)’방법을 제시한다. 또한 실증적 사례에 새 방법론을 적용해보고 토의한다.

• 응용통계연구
• /
• 제21권5호
• /
• pp.727-738
• /
• 2008

변수들 사이 공적분관계의 존재는 불안정한 시계열의 선형결합이 안정적임을 뜻하지만 이 결합이 표본기간 내 발생한 사건에 영향을 받는다면 공적분검정의 기본 가정에 위배되어 검정력은 약해지고 결과는 현실과 차이가 난다. 이러한 관점에 입각하여 본 연구에서는 전체 표본의 일부를 단계적으로 증가시켜 가며 국내 주식시장 장기균형체계를 추정하여 시간가변성 및 안정성을 검정하였으며, 구조적 변화의 발생이 확인된 구간에 대해서는 가변수를 추가하는 방식으로 전 구간에 걸쳐 공적분 벡터를 안정화시키는 방안을 모색하여 보았다.

• Macromolecular Research
• /
• 제15권6호
• /
• pp.553-559
• /
• 2007

Nanocomposite films were prepared by sol-gel synthesis from vanadium triisopropoxide with $poly((oxyethylene)_9$ methacrylate)-graft-poly(dimethyl siloxane), POEM-g-PDMS, producing in situ growth of vanadium oxide within the continuous ion-conducting POEM domains of micro phase-separated graft copolymer. The formation of vanadium oxide was confirmed by wide angle x-ray scattering (WAXS) and Fourier transform infrared (FT-IR) spectroscopy. Small angle x-ray scattering (SAXS) revealed the spatially-selective incorporation of vanadium oxide in the POEM domains. Upon the incorporation of vanadium oxide, the domain periodicity of the graft copolymer monotonously increased from 17.2 to 21.0 nm at a vanadium content 14 v%, above which it remained almost invariant. The selective interaction of vanadium oxide with POEM was further verified by differential scanning calorimetry (DSC) and FT-IR spectroscopy. The nanocomposite films exhibited excellent mechanical properties $(l0^{-5}-10^{-7}dyne/cm^2)$, mostly due to the confinement of vanadium oxide in the POEM chains as well as the interfaces created by the microphase separation of the graft copolymer.

• Journal of the Optical Society of Korea
• /
• 제15권3호
• /
• pp.216-221
• /
• 2011

A diffraction grating is a highly symmetric optical element with a physical structure that is invariant under translational spatial movements. The translational symmetry is reflected in the fields that are diffracted from the grating. Here, we introduce a plane-parallel mirror pair onto the grating, which translates the fields through double reflections, and we describe a method of exploiting the symmetry to enhance the spectral resolution of a diffraction grating beyond the limit that is set by the number of grooves. The mirror pair creates another virtual grating beside the original one, effectively doubling the number of grooves. Addition of more mirror pairs can further increase the effective number of grooves despite the increased complexity and difficulty of experimental implementation. We experimentally demonstrate the spectral linewidth reduction by a factor of four in a neon fluorescence spectrum. Even though the geometrical restriction on the mirror deployment limits our method to a certain range of the whole spectrum, as a practical application example, a bulky spectrometer that is nearly empty inside can be made compact without sacrificing the resolution.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1287-1292
• /
• 2005

This paper presents an effective pattern classification model by designing an artificial neural network based pattern classifiers for face recognition. First, a RGB image inputted from a frame grabber is converted into a HSV image which is similar to the human beings' vision system. Then, the coarse facial region is extracted using the hue(H) and saturation(S) components except intensity(V) component which is sensitive to the environmental illumination. Next, the fine facial region extraction process is performed by matching with the edge and gray based templates. To make a light-invariant and qualified facial image, histogram equalization and intensity compensation processing using illumination plane are performed. The finally extracted and enhanced facial images are used for training the pattern classification models. The proposed H-ART2 model which has the hierarchical ART2 layers and F-LVQ model which is optimized by fuzzy membership make it possible to classify facial patterns by optimizing relations of clusters and searching clustered reference patterns effectively. Experimental results show that the proposed face recognition system is as good as the SVM model which is famous for face recognition field in recognition rate and even better in classification speed. Moreover high recognition rate could be acquired by combining the proposed neural classification models.

• 제어로봇시스템학회논문지
• /
• 제20권11호
• /
• pp.1098-1102
• /
• 2014

Localization of aerial vehicles and map building of flight environments are key technologies for the autonomous flight of small UAVs. In outdoor environments, an unmanned aircraft can easily use a GPS (Global Positioning System) for its localization with acceptable accuracy. However, as the GPS is not available for use in indoor environments, the development of a SLAM (Simultaneous Localization and Mapping) system that is suitable for small UAVs is therefore needed. In this paper, we suggest a vision-based SLAM system that uses vision sensors and an AHRS (Attitude Heading Reference System) sensor. Feature points in images captured from the vision sensor are obtained by using GPU (Graphics Process Unit) based SIFT (Scale-invariant Feature Transform) algorithm. Those feature points are then combined with attitude information obtained from the AHRS to estimate the position of the small UAV. Based on the location information and color distribution, a Gaussian process model is generated, which could be a map. The experimental results show that the position of a small unmanned aircraft is estimated properly and the map of the environment is constructed by using the proposed method. Finally, the reliability of the proposed method is verified by comparing the difference between the estimated values and the actual values.

• 전자공학회논문지SC
• /
• 제40권4호
• /
• pp.251-263
• /
• 2003

본 논문에서는 연속시간과 이산시간에서 파라미터 불확실성을 가지는 선형 시불변 특이시스템에 대한 Η₂제어기 존재조건과 설계방법을 행렬부등식으로 제안한다. 먼저, 연속시간의 경우에는 Η₂제어기가 존재하기 위한 필요충분조건과 설계방법을 선형행렬부등식(linear matrix inequality)으로 제시하고, 이산시간의 경우에는 Η₂제어기가 존재하기 위한 충분조건과 설계방법을 행렬부등식으로 제시한다. 마지막으로 연속시간과 이산시간 각각의 경우에서, 파라미터 불확실성을 고려하여 제시한 조건들을 견실 Η₂제어문제로 확장하고, 간단한 예제를 통해 제시한 조건의 타당성을 검토해 본다.

• Journal of Acupuncture Research
• /
• 제20권4호
• /
• pp.85-92
• /
• 2003

목적 : 허혈시 발생되는 저산소중 상태에서는 세포독성을 유발한다고 알려져 있으나 정확한 기전은 아직 규명되지 않았다. 본 연구에서는 뇌허혈로 인한 세포독성의 기전을 유전자 발현을 통하여 살펴보고자 하였다. 방법 : 본 실험에서는 BV-2 microglia 세포주에 12시간 동안의 저산소 상태에서의 유전자 발현을 분석하기 위하여 마이크로에레이를 시행하였다. 결과 : 저산소 상태에서는 정상에 비하여 cathepsin F, growth factor independent 1, calcitonin/calcitonin-related poly, leucine-rich repeat LGI family membrane, dublecortin, cyclohydrolase 1, Ia-associated invariant chain, carbohydrate kinase-like과 erythrocyte protein band 4.1-like 3 등의 유전자 발현이 3배 이상 증가하였다. 한편 neuronal guanine nucleotide exchange factor, Bcl-2-related ovarian killer protein, chemokine (C-X-C motif) ligand 5, RNA binding motif protein 3, interleukin 2 receptor, alpha chain, crystallin zeta, cytochrome P450 subfamily IV B, asparagine synthetase과 moesin 등의 유전자 발현은 0.2배 이하로 감소하였다. 결론 : 이상의 결과는 저산소중에 관여하는 유전자 및 저산소중과 관련된 뇌경색 등의 질환의 기전을 밝히는데 기초적 자료로 이용될 수 있을 것이다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제17권sup1호
• /
• pp.55-63
• /
• 2016

Smoking and harmful drinking dramatically increase health risks but little is known about their cooccurrence and factors that influence this co-habit, limiting development and implementation of appropriately targeted prevention interventions. This study was conducted among youth aged 10-24 years old in the Chi Linh Demographic - Epidemiological Surveillance System (CHILILAB DESS). The total numbers in the first, second and third rounds in 2006, 2009 and 2013 were 12,406, 10,211, and 7,654, respectively. A random-effects logit model controlling for both time-variant and time-invariant variables was applied to explore factors associated with current smoking, harmful drinking, and occurrence of smoking and harmful drinking together. We found dramatically increasing trends in current smoking, harmful drinking and co-occurrence among youth. Our results indicate similar health problems among youth in peri-urban areas in Vietnam. Demographic characteristics (older age, being male, being unmarried, and having informal work) appeared to be predictors for smoking and drinking behaviour. Besides, peer and family members had significant influence on smoking, whereas having a close-friend who was smoking was the most important variable. The results suggested that smoking and harmful drinking should not be solved with separate, stand-alone interventions but rather with integrated efforts.

• Journal of Computational Design and Engineering
• /
• 제2권1호
• /
• pp.26-37
• /
• 2015

A lot of researchers have been investigating interactive portable projection systems such as a mini-projector. In addition, in exhibition halls and museums, there is a trend toward using interactive projection systems to make viewing more exciting and impressive. They can also be applied in the field of art, for example, in creating shadow plays. The key idea of the interactive portable projection systems is to recognize the user's gesture in real-time. In this paper, a vision-based shadow gesture recognition method is proposed for interactive projection systems. The gesture recognition method is based on the screen image obtained by a single web camera. The method separates only the shadow area by combining the binary image with an input image using a learning algorithm that isolates the background from the input image. The region of interest is recognized with labeling the shadow of separated regions, and then hand shadows are isolated using the defect, convex hull, and moment of each region. To distinguish hand gestures, Hu's invariant moment method is used. An optical flow algorithm is used for tracking the fingertip. Using this method, a few interactive applications are developed, which are presented in this paper.