• 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 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.

• Phonetics and Speech Sciences
• /
• v.3 no.2
• /
• pp.71-78
• /
• 2011

In this paper, we propose an approach which constructs classifier ensembles of various channel compensation and feature enhancement methods. CMN and CMVN are used as channel compensation methods. PCA, kernel PCA, greedy kernel PCA, and kernel multimodal discriminant analysis are used as feature enhancement methods. The proposed ensemble system is constructed with the combination of 15 classifiers which include three channel compensation methods (including 'without compensation') and five feature enhancement methods (including 'without enhancement'). Experimental results show that the proposed ensemble system gives highest average speaker identification rate in various environments (channels, noises, and sessions).

• Korean Journal of Remote Sensing
• /
• v.26 no.6
• /
• pp.693-703
• /
• 2010

High spectral resolution of hyperspectral data enables analysis of complex natural phenomena that is reflected on the data nonlinearly. Although many manifold learning methods have been developed for such problems, most methods do not consider the spatial correlation between samples that is inherent and useful in remote sensing data. We propose a manifold learning method which directly combines the spatial proximity and the spectral similarity through kernel PCA framework. A gain factor caused by spatial proximity is first modelled with a heat kernel, and is added to the original similarity computed from the spectral values of a pair of samples. Parameters are tuned with intelligent grid search (IGS) method for the derived manifold coordinates to achieve optimal classification accuracies. Of particular interest is its performance with small training size, because labelled samples are usually scarce due to its high acquisition cost. The proposed spatial kernel PCA (KPCA) is compared with PCA in terms of classification accuracy with the nearest-neighbourhood classification method.

• The Journal of the Acoustical Society of Korea
• /
• v.30 no.6
• /
• pp.330-336
• /
• 2011

In this paper, we propose a new method to improve the performance of supervector linear kernel SVM (Support Vector Machine) for speaker identification. This method is based on splitting one training datum into several pieces of utterances. We use four different databases for evaluating performance and use PCA (Principal Component Analysis), GKPCA (Greedy Kernel PCA) and KMDA (Kernel Multimodal Discriminant Analysis) for feature enhancement. As a result, the proposed method shows improved performance for speaker identification using supervector linear kernel SVM.

• ETRI Journal
• /
• v.36 no.3
• /
• pp.333-342
• /
• 2014

Most hyper-ellipsoidal clustering (HEC) approaches use the Mahalanobis distance as a distance metric. It has been proven that HEC, under this condition, cannot be realized since the cost function of partitional clustering is a constant. We demonstrate that HEC with a modified Gaussian kernel metric can be interpreted as a problem of finding condensed ellipsoidal clusters (with respect to the volumes and densities of the clusters) and propose a practical HEC algorithm that is able to efficiently handle clusters that are ellipsoidal in shape and that are of different size and density. We then try to refine the HEC algorithm by utilizing ellipsoids defined on the kernel feature space to deal with more complex-shaped clusters. The proposed methods lead to a significant improvement in the clustering results over K-means algorithm, fuzzy C-means algorithm, GMM-EM algorithm, and HEC algorithm based on minimum-volume ellipsoids using Mahalanobis distance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.6
• /
• pp.2302-2316
• /
• 2015

We propose an efficient method of extracting targets within a region of interest in non-homogeneous infrared images by using a principal component analysis (PCA) plane and adaptive Gaussian kernel. Existing approaches for extracting targets have been limited to using only the intensity values of the pixels in a target region. However, it is difficult to extract the target regions effectively because the intensity values of the target region are mixed with the background intensity values. To overcome this problem, we propose a novel PCA based approach consisting of three steps. In the first step, we apply a PCA technique minimizing the total least-square errors of an IR image. In the second step, we generate a binary image that consists of pixels with higher values than the plane, and then calculate the second derivative of the sum of the square errors (SDSSE). In the final step, an iteration is performed until the convergence criteria is met, including the SDSSE, angle and labeling value. Therefore, a Gaussian kernel is weighted in addition to the PCA plane with the non-removed data from the previous step. Experimental results show that the proposed method achieves better segmentation performance than the existing method.

• MALSORI
• /
• no.67
• /
• pp.167-180
• /
• 2008

This paper describes a novel spectral envelope conversion method based on Gaussian mixture model (GMM). The core of this paper is rearranging source feature vectors in input space to the transformed feature vectors in feature space for the better modeling of GMM of source and target features. The quality of statistical modeling is dependent on the distribution and the dimension of data. The proposed method transforms both of the distribution and dimension of data and gives us the chance to model the same data with different configuration. Because the converted feature vectors should be on the input space, only source feature vectors are rearranged in the feature space and target feature vectors remain unchanged for the joint pdf of source and target features using KPCA. The experimental result shows that the proposed method outperforms the conventional GMM-based conversion method in various training environment.

• Korean Journal of Remote Sensing
• /
• v.35 no.6_1
• /
• pp.959-971
• /
• 2019

CNN (Convolutional Neural Network) is one representative deep learning algorithm, which can extract high-level spatial and spectral features, and has been applied for hyperspectral image classification. However, one significant drawback behind the application of CNNs in hyperspectral images is the high dimensionality of the data, which increases the training time and processing complexity. To address this problem, several CNN based hyperspectral image classification studies have exploited PCA (Principal Component Analysis) for dimensionality reduction. One limitation to this is that the spectral information of the original image can be lost through PCA. Although it is clear that the use of PCA affects the accuracy and the CNN training time, the impact of PCA for CNN based hyperspectral image classification has been understudied. The purpose of this study is to analyze the quantitative effect of PCA in CNN for hyperspectral image classification. The hyperspectral images were first transformed through PCA and applied into the CNN model by varying the size of the reduced dimensionality. In addition, 2D-CNN and 3D-CNN frameworks were applied to analyze the sensitivity of the PCA with respect to the convolution kernel in the model. Experimental results were evaluated based on classification accuracy, learning time, variance ratio, and training process. The size of the reduced dimensionality was the most efficient when the explained variance ratio recorded 99.7%~99.8%. Since the 3D kernel had higher classification accuracy in the original-CNN than the PCA-CNN in comparison to the 2D-CNN, the results revealed that the dimensionality reduction was relatively less effective in 3D kernel.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.48 no.5
• /
• pp.7-12
• /
• 2011

In this papwer, we propose a new feature extraction method, named as Class-augmented Kernel Principal Component Analysis (CA-KPCA), which can extract nonlinear features for classification. Among the subspace method that was being widely used for feature extraction, Class-augmented Principal Component Analysis (CA-PCA) is a recently one that can extract features for a accurate classification without computational difficulties of other methods such as Linear Discriminant Analysis (LDA). However, the features extracted by CA-PCA is still restricted to be in a linear subspace of the original data space, which limites the use of this method for various problems requiring nonlinear features. To resolve this limitation, we apply a kernel trick to develop a new version of CA-PCA to extract nonlinear features, and evaluate its performance by experiments using data sets in the UCI Machine Learning Repository.