• The Journal of the Korea Contents Association
• /
• v.11 no.4
• /
• pp.27-32
• /
• 2011

This paper presents a methodology for classifying sleep disturbance using electroencephalogram (EEG) signal at sleep stage 2 and principal component analysis. For extracting initial features, fast Fourier transforms(FFT) were carried out to remove some noise from EEG signal at sleep stage 2. In the second phase, we used principal component analysis to reduction from EEG signal that was removed some noise by FFT to 5 features. In the final phase, 5 features were used as inputs of NEWFM to get performance results. The proposed methodology shows that accuracy rate, specificity rate, and sensitivity were all 100%.

• Journal of Korea Society of Industrial Information Systems
• /
• v.26 no.1
• /
• pp.1-10
• /
• 2021

This study proposed a Long-Short Term Memory network to consider changes in emotion over time, and applied an attention mechanism to give weights to the emotion states that appear at specific moments. We used 32 channel EEG data from DEAP database. A 2-level classification (Low and High) experiment and a 3-level classification experiment (Low, Middle, and High) were performed on Valence and Arousal emotion model. As a result, accuracy of the 2-level classification experiment was 90.1% for Valence and 88.1% for Arousal. The accuracy of 3-level classification was 83.5% for Valence and 82.5% for Arousal.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.10
• /
• pp.105-112
• /
• 2010

In this paper, the bio-signals, such as EEG, ECG were measured with a sensor and their characters were drawn out and analyzed. With results from the analysis, four emotion of rest, concentration, tension and depression were inferred. In order to assess one's emotion, the characteristic vectors were drawn out by applying various ways, including the frequency analysis of the bio-signals like the measured EEG and HRV. RBFN, a neural network of the complex structure of unsupervised and supervised learning, was applied to classify and infer the deducted information. Through experiments, the system suggested in this thesis showed better capability to classify and infer than other systems using a different neural network. As follow-up research tasks, the recognizance rate of the measured bio-signals should be improved. Also, the technology which can be applied to the wired or wireless sensor measuring the bio-signals more easily and to wearable computing should be developed.

• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2002.11a
• /
• pp.461-467
• /
• 2002

Signals reduced from the brain had been considered as a noise that is caused by the stochastic process until 1980. The recent non-linear dynamic theory researches, however, reported that these signals are meaningful and deterministic chaos signals in which they show how the brain deals with various information Since this report, a wide range of researches has been carried out and still in progress. Thus, by using the correlational dimension, one of the non-linear analytical methods, the characteristics of the brain signals can be analyzed. In this thesis, the scent of lavender, which stimulates the olfactory sense, is introduced to measure EEG with the International 10-20 electrode system on 16 channels, and to analyze the interrelationship between the original signals before the stimulation and the changed signals after the stimulation. Finally, the effect of the scent stimulation to the brain is analyzed. The purpose of this thesis is to apply these analyzed results to the computerized mapping of the brain signals and possible ways of specifying the source of the brain signals through various medical applications.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2002.11a
• /
• pp.461-467
• /
• 2002

Signals produced from the brain had been considered as a noise that is caused by the stochastic process until 1980. The recent non-linear dynamic theory researches, however, reported that these signals are meaningful and deterministic chaos signals in which they show how the brain deals with various information Since this report a wide range of researches has been carried out and still in progress. Thus, by using the correlational dimension, one of the non-linear analytical methods, the characteristics of the brain signals can be analyzed. In this thesis, the scent of lavender, which stimulates the olfactory sense, is introduced to measure EEG with the International 10-20 electrode system on 16 channels, and to analyze the interrelationship between the original signals before the stimulation and the changed signals after the stimulation. Finally, the effect of the scent stimulation to the brain is analyzed. The purpose of this thesis is to apply these analyzed results to the computerized mapping of the brain signals and possible ways of specifying the source of the brain signals through various medical applications.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.15 no.3
• /
• pp.16-24
• /
• 2019

Recently, research on emotion recognition based on EEG has attracted great interest from human-robot interaction field. In this paper, we propose a method of labeling using image-based EEG topography instead of evaluating emotions through self-assessment and annotation labeling methods used in MAHNOB HCI. The proposed method evaluates the emotion by machine learning model that learned EEG signal transformed into topographical image. In the experiments using MAHNOB-HCI database, we compared the performance of training EEG topography labeling models of SVM and kNN. The accuracy of the proposed method was 54.2% in SVM and 57.7% in kNN.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.43 no.2
• /
• pp.297-309
• /
• 2019

This study investigated EEG signal, subjective fragrance sensation, and the preference of differently colored cotton knit treated with Citrus unshiu oil containing microcapsules as well as examined their relationships for providing regression models on subjective fragrance preference. Color variables combining 2-level hue (Yellow and Green) and 3-level tone (strong, pale, and grayish) were applied by dyeing prior to microcapsule treatment. We invited 28 female college students aged 20's for EEG signal experiments and subjective fragrance sensations with fragrant knit by rubbing. EEG signals at $mid-{\alpha}$, $fast-{\alpha}$, and $low-{\beta}$ showed significant differences depending on color; Green had more relative power values and grayish tone did more at $low-{\beta}$. Even though subjective sensation showed no significant differences depending on color, some of them such as Fresh, Comfort, and Natural showed significant correlations with EEG signal at $low-{\beta}$, which means that the fragrance sensations of Citrus unshiu fragrance are concerned with attention and alertness for Koreans. Fragrance preference was regressed significantly using some EEG signals and subjective sensation. The results could be utilized to value up fragrant textiles by Citrus unshiu oil.

• Investigative Magnetic Resonance Imaging
• /
• v.8 no.1
• /
• pp.1-8
• /
• 2004

Purpose : Information about electrical activity inside the brain during fMRl scans is very useful in monitoring physiological function of the patient or locating the spatial position of the activated region in the brain. However, many additional noises appear in the EEG signal acquired during the MRI scan. Gradient induced noise is the biggest one among the noises. In this work, we propose a gradient noise reduction method using the independent component analysis (ICA) method. Materials and Methods : We used a 29-channel MR-compatible EEG measurement system and a 3.0 Tesla MRI system. We measured EEG signals on a subject lying inside the magnet during EPI scans. We selectively removed the gradient noise from the measured EEG signal using the ICA method. We compared the results with the ones obtained with conventional averaging method and PCA method. Results : All the noise reduction methods including the averaging and PCA methods were effective in removing the noise in some extent. However, the proposed ICA method was found to be superior to the other methods. Conclusion : Gradient noise in EEG signals acquired during fMRI scans can be effectively reduced by the ICA method. The noise-reduced EEG signal can be used in fMRI studies of epileptic patients or combinatory studies of fMRI and EEG.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2001.11a
• /
• pp.112-116
• /
• 2001

본 연구는 자동차 시뮬레이터에서 필연적으로 발생하는 Sickness와 생리신호 간의 관련성에 대해 알아보고자 하였다. 이를 위해 Driving Simulator에서 중추신경계의 EEG 생리신호를 측정하고 이의 정량적 분석을 통하여 Simulator Sickness가 생리신호에 미치는 영향을 검토하였다. 중추신경계의 뇌파반응은 주행시간의 경과에 따른 특정 경향은 나타나지 않았으나 안정에 비해 delta파, theta파의 증가와 alpha파의 감소경향이 나타났다. 단, theta파는 초기 5분에 증가한 수치가 시간이 지남에 따라 그 비율이 일정하게 감소하는 것으로 나타났다. 또한 Sick그룹과 Non-Sick그룹 및 남녀그룹의 비교결과에 대한 유의한 차이가 인정되었다.

• Journal of Korea Game Society
• /
• v.15 no.2
• /
• pp.105-114
• /
• 2015

We propose a real time user interface that utilizes emotion recognition by physiological signals. To improve the problem that was low accuracy of emotion recognition through the traditional EEG(ElectroEncephaloGram), We developed a physiological signals-based emotion recognition system mixing relative power spectrum values of theta/alpha/beta/gamma EEG waves and autonomic nerve signal ratio of ECG (ElectroCardioGram). We propose both a data map and weight value modification algorithm to recognize six emotions of happy, fear, sad, joy, anger, and hatred. The datamap that stores the user-specific probability value is created and the algorithm updates the weighting to improve the accuracy of emotion recognition corresponding to each EEG channel. Also, as we compared the results of the EEG/ECG bio-singal complex data and single data consisting of EEG, the accuracy went up 23.77%. The proposed interface system with high accuracy will be utillized as a useful interface for controlling the game spaces and smart spaces.