• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2003.04a
• /
• pp.167-170
• /
• 2003

본 연구는 무감독영상해석(Unsupervised Classification)에서 주성분 분석법(Principal Component Analysis)의 응용성을 연구하기 위하여, 주성분 분석법을 K-means, ISODATA 두가지 무감독분류법에 적용하였다. 적용대상지역은 제주도이다. 본 연구에서 주성분 분석 방법중에서 비정규형 주성분 분석방법 (Unstandardized PCA)과 정규형 주성분 분석방법(Standardized PCA) 두가지 경우로 나누어서 각각 연구하였다. 이를 위하여 제주도의 Landsat TM영상과 국토연구원에서 조사한 제주도 식생분류 조사자료와 현장조사 자료 그리고 1/25,000 수치지도를 이용하였다. 그리고 분석된 자료의 정확도를 평가하기 위하여 오차행렬(Error Matrix)을 도입하여 계산하였다. 우선 비정규형 주성분 분석법으로 구한 주성분 영상과 Landsat TM 원래 영상을 오차행렬을 이용하여 제주도의 식생 분류에 각각 적용하였다. 그 결과, K-means 무감독분류법에서는 Landsat TM 자료를 직접 이용한 경우에는 바다와 육상의 분류가 잘 되지 않았으며, 또한 전반적인 영상분류결과가 관측치와 많은 차이를 보였다. 그러나, 주성분 분석법으로 계산된 주성분 영상으로 K-means방법으로 분류 한 결과는 관측치와 잘 일치를 하였다. ISODATA의 경우, Landsat TM 원래영상을 계산하면, K-means으로 분류한 결과보다는 좋은 값을 나타냈으나, 주성분 분석법으로 구한 영상의 계산결과와 비교하면, 주성분 영상으로 구한 분류결과의 정확도가 약 15%정도 높게 나타났다. 정규형 주성분 분석법의 경우를 보면 K-means에서는 Landsat TM원래 자료보다 우수한 결과를 보여주었으나, 비정규형 주성분 분석법으로 계산된 결과보다는 정확도가 다소 떨어지는 단점이 있었고, ISODATA의 경우도 Landsat TM원래 자료보다 약 7%정도의 높은 정확도를 보였으나, 비정규형 영상보다는 약8%정도 낮은 정확도를 보였다. 본 연구에서 주성분 분석법으로 계산된 결과에서 주목되는 것은, 주성분 분석법으로 구한 주성분 영상은 분류방법(K-means, ISODATA, artificial neural networks)에 따라 분류된 결과값이 비슷하게 나타난 반면, Landsat TM원래 자료는 분류방법에 따라 결과값이 많은 차이를 보여 주었다. 그리고 주성분 분석 방법 중에서도 비정규형 주성분 분석법(Unstandardized PCA)이 정규형 주성분 분석법(Standardized PCA)보다 영상분석에서 더 좋은 결과를 보여주는 것으로 나타났다.

• The Korean Journal of Applied Statistics
• /
• v.33 no.1
• /
• pp.61-74
• /
• 2020

Interval-valued data, one type of symbolic data, are observed in the form of intervals rather than single values. Each interval-valued observation has an internal variation. Principal component analysis reduces the dimension of data by maximizing the variance of data. Therefore, the principal component analysis of the interval-valued data should account for the variance between observations as well as the variation within the observed intervals. In this paper, three principal component analysis methods for interval-valued data are summarized. In addition, a new method using a truncated normal distribution has been proposed instead of a uniform distribution in the conventional quantile method, because we believe think there is more information near the center point of the interval. Each method is compared using simulations and the relevant data set from the OECD. In the case of the quantile method, we draw a scatter plot of the principal component, and then identify the position and distribution of the quantiles by the arrow line representation method.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.17 no.7 s.124
• /
• pp.638-648
• /
• 2007

A principal components analysis of the entire median HRIRs in the CIPIC HRTF database reveals that the individual HRIRs can be adequately reconstructed by a linear combination of several orthonormal basis functions. The basis functions represent the inter-individual and inter-elevation variations in median HRIRs. There exist elevation-dependent tendencies in the weights of basis functions, and the basis functions can be ordered according to the magnitude of standard deviation of the weights at each elevation. We propose a HRIR customization method via tuning of the weights of 3 dominant basis functions corresponding to the 3 largest standard deviations at each elevation. Subjective listening test results show that both front-back reversal and vertical perception can be improved with the customized HRIRs.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2001.05a
• /
• pp.74-81
• /
• 2001

얼굴은 내측두(IT: inferotemporal) 영역에 독자적인 처리 공간을 가지고 있는 (Bruce, Desimone, & Gross, 1981; Rolls, 1992) 매우 복잡한 시각 자극이다. 본연구는 이러한 복잡한 얼굴 자극을 구성하고 있는 물리적인 특징들을 추출하여 얼굴을 수치 구조면에서 분석하고 이를 감성 공간과 연결시킬 목적으로 수행되었ㄷ. 이를 위해 본연구에서는 먼저 얼굴 내부에 36개의 특징들 및 특징들 간 관계를 설정하였다. 또한 얼굴 외곽형의 분류를 위해 얼굴 윤곽선 부위에 14개의 특징점을 찍고 코끝에서부터 이들 지점과의 거리를 측정하였다. 사람마다 기본적인 얼굴 14개의 특징점을 찍고 코끝에서부터 이들 지점과의 거리를 측정하였다. 사람마다 기본적인 얼굴 크기가 다르다는 점을 감안하여 이들 특징값들 중 길이값들은 얼굴 좌우폭 또는 얼굴 상하길이를 기주으로 정규화(normalization)되었다. 그런 다음 36개의 얼굴 내부 특징 요소들과 5가지 얼굴 외곽형을 입력값으로 하여 주성분분석(PCA: proncipal component analysis)을 실시하고, 여기서 도출된 다섯 개의 요인점수를 기반으로 5차원 공간을 가정하였다. 이 공간을 대표하는 얼굴을 고루 선정하되 해당 얼굴이 있다고 보기 어려운 영역을 제외하고 평균에 해당하는 얼굴을 추가하여 총 30가지 대표 얼굴 유형을 선정하였다. 선정된 얼굴들에 대해 일차적으로 감성 평가를 실시하여 2차원 감성 공간에 대표 얼굴들을 분포시켰다.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.7
• /
• pp.37-45
• /
• 2013

Context dependent fusion (CDF) is a fusion algorithm that combines multiple outputs from different classifiers to achieve better performance. CDF tries to divide the problem context into several homogeneous sub-contexts and to fuse data locally with respect to each sub-context. CDF showed better performance than existing methods, however, it is sensitive to noise due to the large number of parameters optimized and the innate linearity limits the application of CDF. In this paper, a variant of CDF using support vector machines (SVMs) for fusion and kernel principal component analysis (K-PCA) for context extraction is proposed to solve the problems in CDF, named CDF-SVM. Kernel PCA can shape irregular clusters including elliptical ones through the non-linear kernel transformation and SVM can draw a non-linear decision boundary. Regularization terms is also included in the objective function of CDF-SVM to mitigate the noise sensitivity in CDF. CDF-SVM showed better performance than CDF and its variants, which is demonstrated through the experiments with a landmine data set.

• The KIPS Transactions:PartB
• /
• v.13B no.1 s.104
• /
• pp.53-62
• /
• 2006

Object recognition techniques using principal component analysis are disposed to be decreased recognition rate when lighting change of image happens. The purpose of this thesis is to propose an object recognition technique using new PCA analysis method that discriminates an object in database even in the case that the variation of illumination in training images exists. And the object recognition algorithm proposed here represents more enhanced recognition rate using improved k-Nearest Neighbor. In this thesis, we proposed an object recognition algorithm which creates object space by pre-processing and being learned image using histogram equalization and median filter. By spreading histogram of test image using histogram equalization, the effect to change of illumination is reduced. This method is stronger to change of illumination than basic PCA method and normalization, and almost removes effect of illumination, therefore almost maintains constant good recognition rate. And, it compares ingredient projected test image into object space with distance of representative value and recognizes after representative value of each object in model image is made. Each model images is used in recognition unit about some continual input image using improved k-Nearest Neighbor in this thesis because existing method have many errors about distance calculation.

• Korean Journal of Environment and Ecology
• /
• v.35 no.5
• /
• pp.490-502
• /
• 2021

This study was conducted to analyze the current status of fish fauna and characteristics of the fish community according to the habitat of Woraksan National Park, Korea. The spatially balanced sampling selected 20 stations from major streams of Woraksan National Park, and three surveys were conducted in each season. The physical environments of the habitat were mostly mountain streams (Aa), with large stones and gravels scattered over the stream. The average altitude of the habitat was 304.4 m, and the average depth of water was 40.3 cm, being less than 1 m in most cases, and the river water level was distributed from 3rd to 5th streams. The principal component analysis of the physical environmental factors by habitat showed that the substrate properties differed according to the altitude. The survey identified a total of 2,183 individuals in 16 species belonging to 7 families. The dominant species was Zacco koreanus(86.2%), and the subdominant species was Rhynchocypris oxycephalus(3.8%). Pseudopungtungia tenuicorpa, classified as the endangered wildlife II, was the first endangered legally protected species found in this survey. Analysis of the rank abundance curve model in the fish community showed the Zipf model at 9 out of 20 points, the Lognormal model in 3 points, and the Preemption model in 4 points. The remaining 4 habitats showed only one species and were not analyzed. The canonical correspondence analysis of 20 stations and fish species was performed to understand the characteristics of the fish community according to environmental factors. The fish communities were divided according to differences in habitat environment by the altitude.

• Resources Recycling
• /
• v.26 no.1
• /
• pp.69-77
• /
• 2017

Used small household appliances(small e-waste) consists of a variety of complex materials and components. The small e-waste is mainly composed of plastics and an important potential source of waste plastic. The black plastics, particularly are very difficult to separate by resin type and therefore these are mainly recycled in the form of a mixtures. In the present study, the sorting technologies such as gravity and electro static separation, near-infrared ray(NIR) and IR/Raman optical sorting separation on mixture of black plastics were analyzed and their limitations on sorting process were also investigated. The Laser Induced Breakdown Spectroscopy(LIBS) spectrum of each black plastics was used for identification of black plastics by resin type, and after analyzing the normalization operation, Principal Component Analysis(PCA) was carried out. The spectrum data was optimized through PCA process. In order to improve the identification accuracy and sorting efficiency of black plastics, it is necessary to design a classifier with high efficiency and to improve the performance and reliability of the classifier by applying the field of intelligent algorithms.

• Korean Journal of Cognitive Science
• /
• v.13 no.3
• /
• pp.1-10
• /
• 2002

This study was performed to analyze the physical attributes of the faces and affective words on the fares. For analyzing physical attributes inside of a face, 36 facial features were selected and almost of them were the lengths or distance values. For analyzing facial contour 14 points were selected and the lengths from nose-end to them were measured. The values of these features except ratio values normalized by facial vortical length or facial horizontal length because the face size of each person is different. The principal component analysis (PCA) was performed and four major factors were extracted: 'facial contour' component, 'vortical length of eye' component, 'facial width' component, 'eyebrow region' component. We supposed the five-dimensional imaginary space of faces using factor scores of PCA, and selected representative faces evenly in this space. On the other hand, the affective words on faces were collected from magazines and through surveys. The factor analysis and multidimensional scaling method were performed and two orthogonal dimensions for the affections on faces were suggested: babyish-mature and sharp-soft.