• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.5
• /
• pp.1093-1102
• /
• 2010

Principal component analysis (PCA) is a well-known method for dimension reduction while maintaining most of the variation in data. Although PCA has been applied to many areas successfully, it is sensitive to outliers. Several variants of PCA have been proposed to resolve the problem and, among the variants, robust fuzzy PCA (RF-PCA) demonstrated promising results. RF-PCA uses fuzzy memberships to reduce the noise sensitivity. However, there are also problems in RF-PCA and the convergence property is one of them. RF-PCA uses two different objective functions to update memberships and principal components, which is the main reason of the lack of convergence property. The difference between two functions also slows the convergence and deteriorates the solutions of RF-PCA. In this paper, a variant of RF-PCA, called RF-PCA2, is proposed. RF-PCA2 uses an integrated objective function both for memberships and principal components. By using alternating optimization, RF-PCA2 is guaranteed to converge on a local optimum. Furthermore, RF-PCA2 converges faster than RF-PCA and the solutions found are more similar to the desired solutions than those of RF-PCA. Experimental results also support this.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.4
• /
• pp.15-22
• /
• 2011

Principal component analysis (PCA) is a well-known method for dimensionality reduction and feature extraction while maintaining most of the variation in data. Although PCA has been applied in many areas successfully, it is sensitive to outliers and only valid for Gaussian distributions. Several variants of PCA have been proposed to resolve noise sensitivity and, among the variants, improved robust fuzzy PCA (RF-PCA2) demonstrated promising results. RF-PCA, however, is still a linear algorithm that cannot accommodate non-Gaussian distributions. In this paper, a non-linear algorithm that combines RF-PCA2 and kernel PCA (K-PCA), called improved robust kernel fuzzy PCA (RKF-PCA2), is introduced. The kernel methods make it to accommodate non-Gaussian distributions. RKF-PCA2 inherits noise robustness from RF-PCA2 and non-linearity from K-PCA. RKF-PCA2 outperforms previous methods in handling non-Gaussian distributions in a noise robust way. Experimental results also support this.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.6 no.3
• /
• pp.249-254
• /
• 2005

In this paper, we analysis the recognition performance of PCA/LDA by distance measures. We are adapt to ORL face database with the fourteen distance measures. In case of PCA, it has high performance for the manhattan distance and the weighted SSE distance to face recognition, In case of PCA/LDA, it has high performance for the angle-based distance and the modified SSE distance. Also, PCA/LDA is better than PCA for reduction of dimension. Therefore, the PCA/LDA method and the angle-based distance have the most performance and a few dimension for face recognition with ORL face database.

• Proceedings of the Korean Information Science Society Conference
• /
• /
• pp.538-540
• /
• 2001

PCA는 다변수 데이터 해석법 중 가장 널리 알려진 방법 중 하나로 많은 응용을 가지고 있다. 그런데, PCA는 선형 모델이어서 비선형 구조를 분석하는데 효과적이지 않다. 이를 극복하기 위해서 PCA의 조합을 이용하는 PCA 혼합 모형이 제안되었다. PCA 혼합 모형의 핵심은 구조 선택, 즉 mixture 요소의 수와 PCA 기저의 수의 결정 인데 그의 체계적인 결정 방법이 필요하다. 본 논문에서는 단순화된 PCA 혼합 모형과 이를 위한 효율적인 구조 선택 방법을 제안한다. 각각의 mixture 요소 수에 대해서 모든 PCA 기저를 갖도록 한 상태에서 PCA 혼합 모형의 파라미터를 EM 알고리즘을 써서 결정한다. 최적의 mixture 요소의 수는 오류를 최소로 하는 것으로 결정한다. PCA 기저의 수는 PCA의 정렬성 특성을 이용해서 중요도가 적은 기저부터 하나씩 잘라 내며 오류가 최소로 하는 것으로 결정한다. 제안된 방법은 특히 다차원 데이터의 경우에 EM 학습의 횟수를 많이 줄인다. 인공 데이터에 대한 실험은 제안된 방법이 적절한 모델 구조를 결정한다는 것을 보여준다. 또, 눈 감지에 대한 실험은 제안된 방법이 실용적으로도 유용하다는 것을 보여준다.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.2
• /
• pp.25-31
• /
• 2011

Principal component analysis (PCA) is a well-known method for dimensionality reduction and feature extraction. Although PCA has been applied in many areas successfully, it is sensitive to outliers due to the use of sum-square-error. Several variants of PCA have been proposed to resolve the noise sensitivity and, among the variants, improved robust fuzzy PCA (RF-PCA2) demonstrated promising results. RF-PCA2, however, still can fall into a local optimum due to equal initial membership values for all data points. Another reason comes from the fact that RF-PCA2 is based on sum-square-error although fuzzy memberships are incorporated. In this paper, a variant of RF-PCA2 called RF-PCA3 is proposed. The proposed algorithm is based on the objective function of RF-PCA2. RF-PCA3 augments RF-PCA2 with the objective function of PCA and initial membership calculation using data distribution, which make RF-PCA3 to have more chance to converge on a better solution than that of RF-PCA2. RF-PCA3 outperforms RF-PCA2, which is demonstrated by experimental results.

• MALSORI
• /
• no.66
• /
• pp.105-116
• /
• 2008

In this research, we propose a speaker identification system using a kernel method which is expected to model the non-linearity of speech features well. We have been using principal component analysis (PCA) successfully, and extended to kernel PCA, which is used for many pattern recognition tasks such as face recognition. However, we cannot use kernel PCA for speaker identification directly because the storage required for the kernel matrix grows quadratically, and the computational cost grows linearly (computing eigenvector of $l{\times}l$ matrix) with the number of training vectors I. Therefore, we use greedy kernel PCA which can approximate kernel PCA with small representation error. In the experiments, we compare the accuracy of the greedy kernel PCA with the baseline Gaussian mixture models using MFCCs and PCA. As the results with limited enrollment data show, the greedy kernel PCA outperforms conventional methods.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.6 no.4
• /
• pp.291-295
• /
• 2005

In this paper, we propose a face recognition method using modified 2-D PCA. While the previous PCA method computes the covariance matrix by using one dimensional vectors, the 2-D PCA method computes the covariance matrix by directly using direct two dimensional image, and extracts the feature vectors by solving eigenvalue problem. The proposed method recognizes the faces by applying the modified 2-D PCA to face images and it gets linear transformation matrix using two covariance matrices. The experimental results indicates that the proposed method improved about $1\%$ and achieved more stability in recognition rate than conventional 2-D PCA.

• Proceedings of the Korea Contents Association Conference
• /
• /
• pp.348-351
• /
• 2004

In this paper, we propose the face recognition method using Harr wavelet transform and 2D PCA. While previous PCA computed the covariance matrix by using one dimensional vectors, 2D PCA computed the covarinace matrix by using direct two dimensional image and extracted feature vector by solving eigenvalue problem. To gain the face image having the low dimension and robust property, the proposed method uses wavelet transformation. We apply the LL band image data to 2D PCA for face recognition. The experimental results indicate that our method improves recognition rate than 2D PCA into original image.

• The Korean Journal of Pain
• /
• v.27 no.3
• /
• pp.271-277
• /
• 2014

Background: Postoperative delirium is relatively common. However, the relationship between intravenous patient-controlled analgesia (IV-PCA) and delirium has not been thoroughly investigated. The aim of this study was to evaluate the effects of IV-PCA on the prognosis of postoperative delirium in patients undergoing orthopedic surgery. Methods: Medical records of 129 patients with postoperative delirium were reviewed. Patients were divided into two groups according to whether they used IV-PCA with fentanyl and ketorolac. The IV-PCA group consisted of 73 patients who were managed with IV-PCA; the NO-PCA group consisted of 56 patients who were managed without PCA. Results: Incidences of multiple psychiatric consultations and prolonged delirium were significantly lower in patients using IV-PCA with fentanyl and ketorolac than in those without PCA. Conclusions: We recommend the use of IV-PCA for pain control and management of delirium in patients with postoperative delirium.

• Korean Journal of Food Science and Technology
• /
• v.3 no.2
• /
• pp.105-109
• /
• 1971

Effects of perchloric acid (PCA) and trichloroacetic acid (TCA) on the color generation by orcinol reaction were systematically investigated. When the concentration of PCA on hot acid treatment was varied from 1 through 5 to 10%, and then the concentration of PCA on orcinol reaction was adjusted to 5% of reaction volume, no difference in the color generation was observed between 5 and 10% of PCA, but clearly observed between 1 and 5% PCA. When RNA was treated in 5% hot PCA and then PCA concentrations on orcinol reaction were adjusted to 5% and 10%, respectively, remarkable differences in color generation were observed. When RNA was treated in 10% hot PCA and then PCA concentration on orcinol reaction was adjusted to 5% no difference in color generation between 5% and 10% hot acid treatment was observed. The results show that PCA concentration must be adjusted prior to orcinol reaction.