• 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.

• 한국멀티미디어학회논문지
• /
• 제17권2호
• /
• pp.232-239
• /
• 2014

온라인에서의 글쓰기가 늘어나면서, 기계학습을 통해 이를 분류하는 연구가 늘고 있다. 그럼에도 불구하고 한국어로 작성된 마이크로블로그를 대상으로 한 연구는 많지 않다. 또한 통계적으로 기계학습을 평가한 연구를 찾아보기 힘들다. 본 논문에서는 트위터를 대상으로, 표본을 추출하고, 형태소와 음절을 자질로 사용하여 기계학습에 따라 감정을 분류하였다. 그 결과 약 76%정도 트위터에 포함된 감정이 분류되었다. Support Vector Machine이 Na$\ddot{i}$ve Bayes보다 정확했고, 선형모델도 비구조적인 텍스트 처리에 비선형모델에 상응하는 정확성을 보였다. 또한 형태소가 음절 자질에 비해 높은 정확성을 보이지 않았다.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제31권6호
• /
• pp.817-828
• /
• 2004

본 논문에서는 몽타주 기법과 음영합성 기법을 이용한, 디자이너(예술가)의 감각을 살린 벡터 기반의 얼굴 생성 시스템을 제안한다. 제안하는 시스템은 사진으로부터 얼굴의 특징정보를 추출하여 사람의 얼굴과 유사한 얼굴을 자동으로 생성해 주는 시스템이며, 윤곽선만을 사용하던 기존의 얼굴 생성 시스템과 달리 컬러 기반이며, 음영을 사진으로부터 추출하여 이를 이목구비 이미지와 합성하여 생성하는 방식이다. 따라서 실사형에 좀 더 근접한 얼굴을 생성할 수 있다는 장점을 갖는다. 또, 벡터를 기반으로 하기 때문에 사이즈에 제한 얼이 자유로운 변형이 가능할 뿐만 아니라 디자이너 또는 그림 작가의 느낌을 결과물에 그대로 유지할 수 있도록 한다는 점에서 타 접근방식과의 차별성을 갖는다. 또, 2D 아바타에 자유로운 표정을 적용하는 데에도 쉽게 적용이 가능하다.

• 한국정보전자통신기술학회논문지
• /
• 제5권1호
• /
• pp.13-19
• /
• 2012

본 논문에서는 계층적 삼차원 블록 영상분석을 이용한 폐결절 자동 검출 방법을 제안한다. 제안하는 블록 기반 폐 영상 분석 방법은 크게 두 부분으로 나눌 수 있다. 첫 번째는 블록을 분할하고 분석하고자 하는 대상 블록을 선택하는 과정이며 두 번째는 선택된 분석 대상 블록을 분석하는 과정이다. 첫 번째 과정을 통하여 선택된 분석대상 블록들은 다음 단계인 분석과정을 통해 결절과 비결절로 분리될 수 있다. 분석대상 블록의 중심에 있는 object에서 분석을 위한 형태 특징을 추출 하고, 추출된 형태 특징 값을 Support Vector Machine을 이용하여 결절과 비 결절로 분리한다.

• Structural Engineering and Mechanics
• /
• 제63권1호
• /
• pp.115-124
• /
• 2017

In this paper, a different technique to predict the effects of soil-structure interaction (SSI) on seismic response of building systems is investigated. The technique use a machine learning algorithm called Support Vector Regression (SVR) with technical and analytical results as input features. Normally, the effects of SSI on seismic response of existing building systems can be identified by different types of large data sets. Therefore, predicting and estimating the seismic response of building is a difficult task. It is possible to approximate a real valued function of the seismic response and make accurate investing choices regarding the design of building system and reduce the risk involved, by giving the right experimental and/or numerical data to a machine learning regression, such as SVR. The seismic response of both single-degree-of-freedom system and six-storey RC frame which can be represent of a broad range of existing structures, is estimated using proposed SVR model, while allowing flexibility of the soil-foundation system and SSI effects. The seismic response of both single-degree-of-freedom system and six-storey RC frame which can be represent of a broad range of existing structures, is estimated using proposed SVR model, while allowing flexibility of the soil-foundation system and SSI effects. The results show that the performance of the technique can be predicted by reducing the number of real data input features. Further, performance enhancement was achieved by optimizing the RBF kernel and SVR parameters through grid search.

• 말소리와 음성과학
• /
• 제7권4호
• /
• pp.27-33
• /
• 2015

This paper proposes an audio event classification method using Deep Neural Networks (DNN). The proposed method applies Feed Forward Neural Network (FFNN) to generate event probabilities of ten audio events (dog barks, engine idling, and so on) for each frame. For each frame, mel scale filter bank features of its consecutive frames are used as the input vector of the FFNN. These event probabilities are accumulated for the events and the classification result is determined as the event with the highest accumulated probability. For the same dataset, the best accuracy of previous studies was reported as about 70% when the Support Vector Machine (SVM) was applied. The best accuracy of the proposed method achieves as 79.23% for the UrbanSound8K dataset when 80 mel scale filter bank features each from 7 consecutive frames (in total 560) were implemented as the input vector for the FFNN with two hidden layers and 2,000 neurons per hidden layer. In this configuration, the rectified linear unit was suggested as its activation function.

• 전자공학회논문지
• /
• 제54권3호
• /
• pp.55-62
• /
• 2017

비음수 행렬 분해 기법(non-negative matrix factorization)은 대표적인 부분 영역 기반 표현 기법의 하나로 영상의 부분적인 특징을 나타내는 기저 벡터의 선형 조합으로 영상을 표현하는 기법이다. 본 논문에서는 여러 가지 비음수 행렬 분해 기법을 이용하여 얼굴 영상을 표현하고, 추출된 특징을 기반으로 학습 벡터 양자화를 이용하여 얼굴 인식을 수행하였다. 추출된 각 기법의 기저 벡터를 비교하여 각 기법의 특징을 분석하였다. 또한 NMF 기법들의 인식율 검증을 통해 비음수 행렬 기법의 얼굴 인식에 대한 활용 가능성을 확인하였다.

• 한국멀티미디어학회논문지
• /
• 제20권9호
• /
• pp.1509-1518
• /
• 2017

Although several methods have been used to assess the pain levels, few practical methods for classifying presence or absence of the pain using pattern classifiers have been suggested. The aim of this study is to design an pattern classifier that classifies the presence or absence of the pain using electrocardiogram (ECG). We measured the ECG signal from 10 subjects with the painless state and the pain state(Induced by mechanical stimulation). The 10 features of heart rate variability (HRV) were extracted from ECG - MeanRRI, SDNN, rMSSD, NN50, pNN50 in the time domain; VLF, LF, HF, Total Power, LF/HF in the frequency domain; and we used the features as input vector of the pattern classifier's artificial neural network (ANN) / support vector machine (SVM) for classifying the presence or absence of the pain. The study results showed that the classifiers using ANN / SVM could classify the presence or absence of the pain with accuracies of 81.58% / 81.84%. The proposed classifiers can be applied to the objective assessment of pain level.

• 응용통계연구
• /
• 제25권5호
• /
• pp.829-835
• /
• 2012

When the input features are generated by factors in a classification problem, it is more meaningful to identify important factors, rather than individual features. The $F_{\infty}$-norm support vector machine(SVM) has been developed to perform automatic factor selection in classification. However, the $F_{\infty}$-norm SVM may suffer from estimation inefficiency and model selection inconsistency because it applies the same amount of shrinkage to each factor without assessing its relative importance. To overcome such a limitation, we propose the adaptive $F_{\infty}$-norm ($AF_{\infty}$-norm) SVM, which penalizes the empirical hinge loss by the sum of the adaptively weighted factor-wise $L_{\infty}$-norm penalty. The $AF_{\infty}$-norm SVM computes the weights by the 2-norm SVM estimator and can be formulated as a linear programming(LP) problem which is similar to the one of the $F_{\infty}$-norm SVM. The simulation studies show that the proposed $AF_{\infty}$-norm SVM improves upon the $F_{\infty}$-norm SVM in terms of classification accuracy and factor selection performance.

• Structural Monitoring and Maintenance
• /
• 제6권2호
• /
• pp.125-145
• /
• 2019

Bridge scour is one of the predominant causes of bridge failure. Current climate deterioration leads to increase of flooding frequency and severity and thus poses a higher risk of bridge scour failure than before. Recent studies have explored extensively the vibration-based scour monitoring technique by analyzing the structural modal properties before and after damage. However, the state-of-art of this area lacks a systematic approach with sufficient robustness and credibility for practical decision making. This paper attempts to develop a data-driven methodology for bridge scour monitoring using support vector machines. This study extracts features from the bridge dynamic responses based on a generic sensitivity study on the bridge's modal properties and selects the features that are significantly contributive to bridge scour detection. Results indicate that the proposed data-driven method can quantify the bridge scour damage with satisfactory accuracy for most cases. This paper provides an alternative methodology for bridge scour evaluation using the machine learning method. It has the potential to be practically applied for bridge safety assessment in case that scour happens.