• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.1
• /
• pp.196-204
• /
• 2008

In this paper, we propose an efficient method to classify human facial impression using frontal face image. The features that represent the shape of eye, jaw and face are used. The proposed method employs PCA, LDA and SVM in series. PCA is used to project the feature space to a low dimensional subspace. LDA produces well separated classes in a low dimensional subspace even under severe variation. This results in good discriminating power for classification. SVM is used to classify the data. Human face has been classified for 8 facial impressions. The experiments have been performed for many face images, and show encouraging result.

• Korean Journal of Optics and Photonics
• /
• v.31 no.3
• /
• pp.125-133
• /
• 2020

An innovative method for separating overlapping latent fingerprints, using laser-induced plasma spectroscopy (LIPS) combined with multivariate analysis, is reported in the current study. LIPS provides the capabilities of real-time analysis and high-speed scanning, as well as data regarding the chemical components of overlapping fingerprints. These spectra provide valuable chemical information for the forensic classification and reconstruction of overlapping latent fingerprints, by applying appropriate multivariate analysis. This study utilizes principal-component analysis (PCA) and partial-least-squares (PLS) techniques for the basis classification of four types of fingerprints from the LIPS spectra. The proposed method is successfully demonstrated through a classification example of four distinct latent fingerprints, using discrimination such as soft independent modeling of class analogy (SIMCA) and partial-least-squares discriminant analysis (PLS-DA). This demonstration develops an accuracy of more than 85% and is proven to be sufficiently robust. In addition, by laser-scanning analysis at a spatial interval of 125 ㎛, the overlapping fingerprints were separated as two-dimensional forms.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.18 no.6
• /
• pp.642-653
• /
• 2008

This study deals with modeling of head-related transfer functions(HRTFs) using principal components analysis(PCA) in the time and frequency domains. Four PCA models based on head-related impulse responses(HRIRs), complex-valued HRTFs, augmented HRTFs, and log-magnitudes of HRTFs are investigated. The objective of this study is to compare modeling performances of the PCA models in the least-squares sense and to show the theoretical relationship between the PCA models. In terms of the number of principal components needed for modeling, the PCA model based on HRIR or augmented HRTFs showed more efficient modeling performance than the PCA model based on complex-valued HRTFs. The PCA model based on HRIRs in the time domain and that based on augmented HRTFs in the frequency domain are shown to be theoretically equivalent. Modeling performance of the PCA model based on log-magnitudes of HRTFs cannot be compared with that of other PCA models because the PCA model deals with log-scaled magnitude components only, whereas the other PCA models consider both magnitude and phase components in linear scale.

• Korean Chemical Engineering Research
• /
• v.45 no.1
• /
• pp.87-92
• /
• 2007

Recently, developments of process monitoring system in order to detect and diagnose process abnormalities has got the spotlight in process systems engineering. Normal data obtained from processes provide available information of process characteristics to be used for modeling, monitoring, and control. Since modern chemical and environmental processes have high dimensionality, strong correlation, severe dynamics and nonlinearity, it is not easy to analyze a process through model-based approach. To overcome limitations of model-based approach, lots of system engineers and academic researchers have focused on statistical approach combined with multivariable analysis such as principal component analysis (PCA), partial least squares (PLS), and so on. Several multivariate analysis methods have been modified to apply it to a chemical process with specific characteristics such as dynamics, nonlinearity, and so on.This paper discusses about missing value estimation and sensor fault identification based on process variable reconstruction using dynamic PCA and canonical variate analysis.

• Journal of Broadcast Engineering
• /
• v.19 no.1
• /
• pp.64-74
• /
• 2014

The primary and ambient signal decomposition of a stereo sound is a key step to the stereo upmix. The principal component analysis (PCA) is one of the most widely used methods of primary-ambient signal decomposition. However, previous PCA-based decomposition algorithms assume that stereo sound sources are only amplitude-panned without any consideration of phase difference. So it occurs some performance degradation in case of live recorded stereo sound. In this paper, we propose a new PCA-based stereo decomposition algorithm that can consider the phase difference between the channel signals. The proposed algorithm overcomes limitation of conventional signal model using PCA with phase alignment. The phase alignment is realized by using inter-channel phase difference (IPD) which is widely used in parametric stereo coding. Moreover, Enhanced Modified PCA(EMPCA) is combined to solve the problem of conventional PCA caused by Primary to Ambient energy Ratio(PAR) and panning angle dependency. The simulation results are presented to show the improvements of the proposed algorithm.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.7
• /
• pp.519-527
• /
• 2003

Speech is much influenced by the existence of outliers which are introduced by such an unexpected happenings as additive background noise, change of speaker's utterance pattern and voice detection errors. These kinds of outliers may result in severe degradation of speaker recognition performance. In this paper, we proposed the GMM based on robust principal component analysis (RPCA-GMM) using M-estimation to solve the problems of both ouliers and high dimensionality of training feature vectors in speaker identification. Firstly, a new feature vector with reduced dimension is obtained by robust PCA obtained from M-estimation. The robust PCA transforms the original dimensional feature vector onto the reduced dimensional linear subspace that is spanned by the leading eigenvectors of the covariance matrix of feature vector. Secondly, the GMM with diagonal covariance matrix is obtained from these transformed feature vectors. We peformed speaker identification experiments to show the effectiveness of the proposed method. We compared the proposed method (RPCA-GMM) with transformed feature vectors to the PCA and the conventional GMM with diagonal matrix. Whenever the portion of outliers increases by every 2%, the proposed method maintains almost same speaker identification rate with 0.03% of little degradation, while the conventional GMM and the PCA shows much degradation of that by 0.65% and 0.55%, respectively This means that our method is more robust to the existence of outlier.

• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06c
• /
• pp.191-194
• /
• 2011

문서를 이루는 단어들의 빈도수가 지수법칙(power law)를 따른다는 지프의 법칩(Zipf's law)이 있다. 이러한 단어분포를 고려하여 문서의 토픽을 찾아내는 기계학습법이 디리쉴레 프로세스(Dirichlet process) 이다. 이를 발전시켜서 데이터의 잠재 요인(latent factor)들을 베이즈 확률모델에 기반한 샘플링 바탕으로 찾는 방법이 인디언 뷔페 과정(Indian buffet process) 이다. 우리는 25가지의 특징(feature)들에 대한 점수(rating)들이 볼드(blood oxygen dependent level) 신호와 함께 주어지는 PBAIC 2007 데이터에 주성분 분석법(principal component analysis)를 적용했다. PBAIC 2007 데이터는 비디오 게임을 수행하며 기능적뇌영상(functional magnetic resonance imaging, fMRI) 촬영을 하여 얻어진 공개데이터이다. 우리의 연구에서는 주성분 분석법을 이용하여 10개의 독립 성분(independent component)들을 찾았다. 그리고 1.75초 마다 촬영된 BOLD 신호와 10개의 고유벡터(eigenvector)들간의 내적을 취하여 가중치(weight)를 구하였다. 성분들의 가중치를 낮은 순서로 정렬함으로써 각 시간마다 주도적으로 영향을 미치는 성분들을 알아낼 수 있었다.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2008.04a
• /
• pp.920-927
• /
• 2008

This study deals with modeling of Head-Related Transfer Functions (HRTFs) using Principal Components Analysis (PCA) in the time and frequency domains. Four PCA models based on Head-Related Impulse Responses (HRIRs), complex-valued HRTFs, augmented HRTFs, and log-magnitudes of HRTFs are investigated. The objective of this study is to compare modeling performances of the PCA models in the least-squares sense and to show the theoretical relationship between the PCA models. In terms of the number of principal components needed for modeling, the PCA model based on HRIR or augmented HRTFs showed more efficient modeling performance than the PCA model based on complex-valued HRTFs. The PCA model based on HRIRs in the time domain and that based on augmented HRTFs in the frequency domain are shown to be theoretically equivalent. Modeling performance of the PCA model based on log-magnitudes of HRTFs cannot be compared with that of other PCA models because the PCA model deals with log-scaled magnitude components only, whereas the other PCA models consider both magnitude and phase components in linear scale.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.6
• /
• pp.64-69
• /
• 2008

This paper discusses low resolution face recognition using the photon-counting linear discriminant analysis (LDA). The photon-counting LDA asymptotically realizes the Fisher criterion without dimensionality reduction since it does not suffer from the singularity problem of the fisher LDA. The linear discriminant function for optimal projection is determined in high dimensional space to classify unknown objects, thus, it is more efficient in dealing with low resolution facial images as well as conventional face distortions. The simulation results show that the proposed method is superior to Eigen face and Fisher face in terms of the accuracy and false alarm rates.

• Journal of Sericultural and Entomological Science
• /
• v.31 no.2
• /
• pp.102-112
• /
• 1989

Principal component analysis and cluster analysis were performed on the nine quantitative characters of the one hundred and forty eight silkworm genetic stocks. The six major quantitative characters such as cocoon yield, cocoon weight, cocoon shell weight, cocoon shell percentage, larval period of the 5th instar silkworm, and total larval period showed significantly positive correlation between them. The first three principal components extracted form the initial nine variables by principal component analysis accounted for about eighty percent of original information. The first and second principal components were characterized as factors related to silk productivity, and cocoon productivity, respectively. On the basis of multivariate analysis using city block distance determined from the first three principal components to measure the phenotypic diversity, the one hundred and forty eight silkworm genetic stocks could be clustered into seven varietal groups, and the phenotypic diversity between the varietal groups was partly related to their geographical origins. Among 7 varietal group, group II and IV revealed higher silk and cocoon productivity.