• 대한공간정보학회지
• /
• 제13권3호
• /
• pp.33-40
• /
• 2005

위성영상에서 실제 지표면의 형태와 지상물체를 구분하여 분류하는 것은 원격탐사의 중요한 목적중의 하나이다. 다중분광영상을 이용한 분류는 일반적인 토지피복도의 제작에 이용되어지고 있으며 영상분류의 방법에는 많은 이론들이 사용되어지고 있다. 본 연구는 대구 달성군 지역의 IKONOS 영상을 MLC(Maximum Likelihood Classification), ANN(Artificial neural network), SVM(Support Vector Machine), Naive Bayes 분류기법들을 이용하여 각각의 분류정확도를 비교 분석하였다. 또한 PCA/ICA 전처리 과정을 거친 분류기법들 결과와, Boosting 알고리즘 과정을 거친 후의 결과를 비교하였다. 본 연구의 목적은 적절한 전처리과정과 분류기법을 수행함으로써 가장 효율적인 지형분류 방법을 획득하는데 그 목적이 있다.

• Journal of Advanced Research in Ocean Engineering
• /
• 제2권4호
• /
• pp.202-208
• /
• 2016

In this paper, we propose novel seabed sediment classification algorithm using feature obtained by continuous wavelet transform (CWT). Contrast to previous researches using direct reflection coefficient of seabed which is function of frequency and is highly influenced by sediment types, we develop an algorithm using both direct reflection signal and backscattering signal. In order to obtain feature vector, we employ CWT of the signal and obtain histograms extracted from local binary patterns of the scalogram. The proposed algorithm also adopts principal component analysis (PCA) to reduce dimension of the feature vector so that it requires low computational cost to classify seabed sediment. For training and classification, we adopts K-means clustering algorithm which can be done with low computational cost and does not require prior information of the sediment. To verify the proposed algorithm, we obtain field data measured at near Jeju island and show that the proposed classification algorithm has reliable discrimination performance by comparing the classification results with actual physical properties of the sediments.

• Journal of Electrical Engineering and Technology
• /
• 제9권3호
• /
• pp.1060-1071
• /
• 2014

Signals of the Electroencephalogram (EEG) can reflect the electrical background activity of the brain generated by the cerebral cortex nerve cells. This has been the mostly utilized signal, which helps in effective analysis of brain functions by supervised learning methods. In this paper, an approach for improving the accuracy of EEG signal classification is presented to detect epileptic seizures. Moreover, Independent Component Analysis (ICA) is incorporated as a preprocessing step and Short Time Fourier Transform (STFT) is used for denoising the signal adequately. Feature extraction of EEG signals is accomplished on the basis of three parameters namely, Standard Deviation, Correlation Dimension and Lyapunov Exponents. The Artificial Neural Network (ANN) is trained by incorporating Levenberg-Marquardt(LM) training algorithm into the backpropagation algorithm that results in high classification accuracy. Experimental results reveal that the methodology will improve the clinical service of the EEG recording and also provide better decision making in epileptic seizure detection than the existing techniques. The proposed EEG signal classification using feed forward Backpropagation Neural Network performs better than to the EEG signal classification using Adaptive Neuro Fuzzy Inference System (ANFIS) classifier in terms of accuracy, sensitivity, and specificity.

• Communications for Statistical Applications and Methods
• /
• 제21권1호
• /
• pp.61-68
• /
• 2014

We propose binary classification methods by modifying logistic regression classification. We use variable selection procedures instead of original variables to select the principal components. We describe the resulting classifiers and discuss their properties. The performance of our proposals are illustrated numerically and compared with other existing classification methods using synthetic and real datasets.

• 한국음향학회지
• /
• 제38권6호
• /
• pp.687-696
• /
• 2019

기존의 지진파 분류 특징은 강진에 초점이 맞추어져 있어서 미소지진과 같은 지진파는 다소 적합하지 않다. 본 연구에서는 강진과 더불어 미소지진, 인공지진, 잡음 분류에 적합한 특징 추출을 위해 주파수-시간 공간 내에서 히스토그램과 주성분 기반 특징 추출방법을 제안한다. 제안된 방법은 지진파의 주파수 관련 정보와 시간 관련 정보를 결합하는 방법을 적용한 히스토그램 기반 특징 추출방법과 주성분 기반 특징 추출방법을 이용하여 지진(강진, 미소지진, 인공지진)과 잡음, 미소지진과 잡음, 미소지진과 인공지진을 이진 분류한다. 2017년~2018년 최근 국내지진 자료와 분류 성능을 토대로 제안한 특징 추출방식의 효용성을 비교 평가한다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.496-498
• /
• 2003

Signal is composed of the independent components that can describe itself. These components can distinguish itself from any other signals and be extracted by analysis itself. This algorithm is called Independent Component Analysis (ICA) and image signal is considered as linear combination of independent components and features that is the weighted vector of independent component. This algorithm is already used in order to extract the good feature for image classification and very effective In this paper, we'll explain the method of stereo matching using independent component analysis and show the experimental result.

• Journal of the Korean Wood Science and Technology
• /
• 제45권2호
• /
• pp.202-212
• /
• 2017

This study was to establish the interrelation between chemical compositions and near infrared (NIR) spectra for the classification on distinguishability of domestic gymnosperms. Traditional wet chemistry methods and infrared spectral analyses were performed. In chemical compositions of five softwood species including larch (Larix kaempferi), red pine (Pinus densiflora), Korean pine (Pinus koraiensis), cypress (Chamaecyparis obtusa), and cedar (Cryptomeria japonica), their extractives and lignin contents provided the major information for distinction between the wood species. However, depending on the production region and purchasing time of woods, chemical compositions were different even though in same species. Especially, red pine harvested from Naju showed the highest extractive content about 16.3%, whereas that from Donghae showed about 5.0%. These results were expected due to different environmental conditions such as sunshine amount, nutrients and moisture contents, and these phenomena were also observed in other species. As a result of the principal component analysis (PCA) using NIR between five species (total 19 samples), the samples were divided into three groups in the score plot based on principal component (PC) 1 and principal component (PC) 2; group 1) red pine and Korean pine, group 2) larch, and group 3) cypress and cedar. Based on the chemical composition results, it was concluded that extractive content was highly relevant to wood classification by NIR analysis.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권1호
• /
• pp.12-18
• /
• 2013

Motor imagery classification in electroencephalography (EEG)-based brain-computer interface (BCI) systems is an important research area. To simplify the complexity of the classification, selected power bands and electrode channels have been widely used to extract and select features from raw EEG signals, but there is still a loss in classification accuracy in the state-of- the-art approaches. To solve this problem, we propose a discriminative feature extraction algorithm based on power bands with principle component analysis (PCA). First, the raw EEG signals from the motor cortex area were filtered using a bandpass filter with ${\mu}$ and ${\beta}$ bands. This research considered the power bands within a 0.4 second epoch to select the optimal feature space region. Next, the total feature dimensions were reduced by PCA and transformed into a final feature vector set. The selected features were classified by applying a support vector machine (SVM). The proposed method was compared with a state-of-art power band feature and shown to improve classification accuracy.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권3호
• /
• pp.186-199
• /
• 2013

We pose pattern classification as a density estimation problem where we consider mixtures of generative models under partially labeled data setups. Unlike traditional approaches that estimate density everywhere in data space, we focus on the density along the decision boundary that can yield more discriminative models with superior classification performance. We extend our earlier work on the recursive estimation method for discriminative mixture models to semi-supervised learning setups where some of the data points lack class labels. Our model exploits the mixture structure in the functional gradient framework: it searches for the base mixture component model in a greedy fashion, maximizing the conditional class likelihoods for the labeled data and at the same time minimizing the uncertainty of class label prediction for unlabeled data points. The objective can be effectively imposed as individual mixture component learning on weighted data, hence our mixture learning typically becomes highly efficient for popular base generative models like Gaussians or hidden Markov models. Moreover, apart from the expectation-maximization algorithm, the proposed recursive estimation has several advantages including the lack of need for a pre-determined mixture order and robustness to the choice of initial parameters. We demonstrate the benefits of the proposed approach on a comprehensive set of evaluations consisting of diverse time-series classification problems in semi-supervised scenarios.

• 한국지형학회지
• /
• 제19권3호
• /
• pp.37-49
• /
• 2012

이 논문의 목적은 간석지 표층 퇴적상 분류를 목적으로 다변량 크리깅을 기반으로 고해상도 원격탐사 자료와 현장 조사 자료를 결합하는 방법론을 제안하는데 있다. 퇴적물 성분에 따라 미리 범주화시킨 퇴적물 자료를 사용하여 원격탐사 자료를 분류하는 기존 방법론과 달리 현장 조사 자료와 원격탐사 자료를 이용하여 퇴적물 성분별 분포도를 제작한 후에 최종 단계에서 범주화 시키는 분류 방법론을 제안하였다. 퇴적물 성분별 분포도 제작 과정에서 현장 조사 자료와 원격탐사 자료의 결합을 위해 다변량 크리깅 기법인 회귀 크리깅 기법을 이용하였다. 우선 현장조사 자료의 모래, 실트, 점토 성분별로 고해상도 원격탐사 자료의 분광 정보와 회귀 분석을 수행하여, 각 성분별 경향 성분을 추출하였다. 그리고 현장 조사 자료 위치에서 잔차를 계산한 후에, 잔차에 대해 크리깅을 적용하여 잔차분포도를 얻게 된다. 이후 성분별 경향 성분과 잔차 성분을 합하여 성분별 비율 분포도를 작성한 후에 최종 단계에서 퇴적상 분류를 수행하게 된다. 제안 기법의 적용성 평가를 위해 바람아래 간석지를 대상으로 고해상도 KOMPSAT-2 자료를 이용한 사례 연구를 수행하였다. 사례 연구를 통해 제안 기법이 기존 분류 방법에 비해 상대적으로 높은 분류 정확도를 나타내었으며, 특히 세립질 퇴적물 분류에 더 우수한 것으로 나타났다. 따라서 제안 기법은 원격탐사 자료를 이용한 간석지 표층 퇴적상 분류에 유용하게 사용될 수 있을 것으로 기대된다.