• 대한전자공학회논문지SP
• /
• 제45권4호
• /
• pp.100-106
• /
• 2008

본 논문에서는 다양한 잡음 환경에서 음성의 통계적 모델에 기반한 음성 검출기의 성능향상을 위해 PSFM (Power Spectral Flatness Measure)을 이용하여 실시간으로 변별적 가중치 학습 (Discriminative Weight Training) 기반의 최적화된 우도비 테스트 (Likelihood Ratio Test, LRT)를 제안한다. 먼저, 기존의 통계모델기반의 음성 검출기를 분석하고, 이를 기반으로 MCE (Minimum Classification Error)방법을 도입하여 도출한 각 주파수 채널별 가중치를 PSFM 값에 기반하여 실시간 매 프레임마다 다른 가중치를 적용한 우도비 기반의 음성 검출 결정법을 제시한다. 제안된 알고리즘은 다양한 잡음 환경에서 기존에 제시된 음성 검출기와 비교하였으며, 우수한 성능을 보인다.

• 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.

• 전자공학회논문지
• /
• 제51권5호
• /
• pp.197-206
• /
• 2014

운동심상(Motor imagery) 기반의 뇌-컴퓨터 인터페이스(Brain-computer Interface)는 주로 뇌전도(Electroencephalography, EEG)를 이용하여 사용자의 자발적인 운동 의지를 읽는 기술로 최근 주목받고 있다. 이 중에서도 피실험자의 운동 의지를 정확히 해석하기 위해 감각운동 영역(sensorimotor area)의 일부분에서 나타나는 ${\mu}$-대역(8-13Hz)의 전위 감소 현상인 event related desynchronization(ERD)을 분석하는 연구가 많이 진행되고 있다. 하지만 EEG는 공간 해상도가 낮고 사용자에 따라 ERD가 발생하는 주파수 대역이 다소 차이가 있어 추정에 어려움이 있다. 이에 대한 개선 방법의 하나로서 공간 필터를 구현하는 common spatial pattern (CSP)과 필터 뱅크(filter bank)를 결합한 형태인 discriminative filter bank common spatial pattern(DFBCSP)이 제안되었다. 그러나 DFBCSP는 EEG 신호의 평균 전력(power)의 Fisher ratio를 이용하여 사용자에 따른 효과적인 주파수 대역을 포함하는 discriminative filter bank(DFB)를 구성하여 분류 정확도를 향상시켰지만 ERD의 공간 패턴이 나타나는 적절한 필터를 선택하지 않는 경우가 발생한다. 이러한 문제를 해결하기 위해 본 논문에서는 EEG 신호의 평균전력 대신 CSP의 특성 벡터를 이용하여 DFB를 구성하는 방법을 제안한다. 기존의 방법과 제안한 방법의 필터 선택 결과와 분류 정확도 분석을 통해 CSP 특성 벡터가 DFB 구성에 더욱 효과적임을 보인다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권1호
• /
• pp.305-326
• /
• 2019

Discriminative correlation filter (DCF) based tracking algorithms have recently shown impressive performance on benchmark datasets. However, amount of recent researches are vulnerable to heavy occlusions, irregular deformations and so on. In this paper, we intend to solve these problems and handle the contradiction between accuracy and real-time in the framework of tracking-by-detection. Firstly, we propose an innovative strategy to combine the template and color-based models instead of a simple linear superposition and rely on the strengths of both to promote the accuracy. Secondly, to enhance the discriminative power of the learned template model, the spatial regularization is introduced in the learning stage to penalize the objective boundary information corresponding to features in the background. Thirdly, we utilize a discriminative multi-scale estimate method to solve the problem of scale variations. Finally, we research strategies to limit the computational complexity of our tracker. Abundant experiments demonstrate that our tracker performs superiorly against several advanced algorithms on both the OTB2013 and OTB2015 datasets while maintaining the high frame rates.

• 아동학회지
• /
• 제13권2호
• /
• pp.203-216
• /
• 1992

The purpose of this study was to investigate the utility of the Luria-Nebraska Battery-Children's Revision (LNNB-CR) in the clinical situation in Korea: LNNB-CR was applied clinically in order to examine how well it differentiates brain damaged children from normals. Subjects were 30 children with the average age of 10 years. Among them 15 were diagnosed as the 'brain damaged' group, and the remaining 15 were normal. All subjects of the two groups were tested on all items in accordance with LNNB-CR manual. Data were analyzed by percentage, correlation coefficient, and t-test. The results were (1) the degree of consistency between testers averaged 97.2％ which confirmed the stability of the scoring system. (2) Split-half reliability was ranged from .75 on the Tactile Scale (C3) to .95 on the Memory Scale (C10). Thus, consistency of items within the scales appeared high. (3) Internal consistency reliability ranged from .74 on the Visual Scale (C4) to .98 on the Reading Scale (C9). Thus, the homogenity of items within the scales appeared high. (4) In the diagnostic discriminative power test between the two groups, both individual scales and total scores showed significant differences at the level of p<.001. (5) The discriminative power test between two groups on all items showed significant differences at the level of p<.05 or better in 126(85％) out of 149 items. These results are supportive of the diagnostic utility of the application of LNNB-CR to the clinical situation in Korea. However, extensive additional research is needed in order to prove its worth.

• 경영과정보연구
• /
• 제25권
• /
• pp.203-222
• /
• 2008

This thesis examines whether accounting information-earnings and book values-has the value relevance in the KOSDAQ. The study is motivated by previous studies which have examined the value relevance of accounting information. Prior researches have focused on KSE(Korea Stock Exchange). But, prior researches have not examined the value relevance of accounting information in KOSDAQ. So, this study examined the value relevance of accounting information which is disclosed by firms on KOSDAQ and whether accounting information between firms on KOSDAQ and KSE has the discriminative value relevance, underlying the expectation that KOSDAQ firms will have higher future profitability than KSE firms. In other words, book-value multiples of KOSDAQ firms is higher than book-value multiples of KSE firms and earnings multiples of KOSDAQ firms is lower than earnings multiples of KSE firms. The value relevance of accounting information is examined by a valuation framework presented by Ohlson(1995), which expresses the stock-price as a function of both earnings and book values of equity. The results indicate that accounting information of KOSDAQ has significant explanatory power for stock price over the 2005-2007 period. KOSDAQ firm are divided by Venture firms and Small to Mid size firms. KOSDAQ Venture firms have the discriminative value relevance, compared with KSE firms. But, KOSDAQ Small to Mid size firms have not the discriminative value relevance, compared with KSE firms.

• 대한영양사협회학술지
• /
• 제15권4호
• /
• pp.364-382
• /
• 2009

The purposes of this study were to develop the standard indicators to evaluate the food and nutrition systems in hospitals and to test the validity of those items scientifically. The results were as follows: First, the conceptual validity was examined with recognition degrees of importance from the hospital nutrition department managers. All of the hospital nutrition department's operation evaluation standards and the indicators' conceptual validity tested were in the range of 3.71~4.93 out of 5.0, and the mean score was 4.36. Therefore, the conceptual validity was verified. Second, to verify the factor validity of the items of the standards and indicators for the hospital nutrition department's operation evaluation, the standards and indicators were analyzed as key-factors. Key-factor analysis after vertical rotation showed that four factors appeared and were composed of (a) facilities management, (b) sanitation management, (c) operation & foodservice management, and (d) nutrition management. Third, the reliability of the standards and indicators for the hospital nutrition department's operation evaluation was analyzed and resulted in a score of 0.98, which showed good internal consistency. Fourth, the discriminative power of each item of the standards for the hospital nutrition department's operation evaluation was tested by checking the differences between groups with first quartile and forth quartile of total evaluation scores. The indicators having low distinction power were modified into obligatory items or eliminated for better differentiation.

• 한국보육지원학회지
• /
• 제18권4호
• /
• pp.1-16
• /
• 2022

Objective: The purpose of this study is to test the psychometric properties of the Maternal Emotion Coaching Questionnaire (MECQ, Lim et al., 2018) in order to measure emotion coaching in mothers of preschoolers. Methods: A total of 316 preschoolers and their mothers participated in this study. Maternal emotion coaching was assessed by self-report and child emotion regulation ability was evaluated by the teacher. Data were analyzed with chi-square tests, reliability analysis, confirmatory factor analysis, latent profile analysis, and F-test. Results: Each item of the MECQ showed proper discriminative power. The MECQ and each subscale demonstrated adequate internal consistency and split-half reliability. Evidence of construct validity was provided by confirmatory factor analysis. The five-factor model including maternal attention, awareness, acceptance, empathy, and guidance showed a good fit. Results of the latent profile analysis revealed three profiles of emotion coaching: excellent, good, and poor. Preschoolers with mothers in the poor coaching profile showed significantly lower emotion regulation ability compared to those in the excellent or good coaching profiles, which suggested discriminative validity of the MECQ. Conclusion/Implications: The MECQ presents a reliable and valid tool to assess emotion coaching in mothers of preschool children and can thus be effectively used for mothers of preschoolers.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2002년도 11월 학술대회지
• /
• pp.115-118
• /
• 2002

Neural networks have been known to have great discriminative power in pattern classification problems. In this paper, the multilayer perceptron neural networks are employed to automatically detect laryngeal pathology in speech. Also new feature parameters are introduced which can reflect the periodicity of speech and its perturbation. These parameters and cepstral coefficients are used as input of the multilayer perceptron neural networks. According to the experiment using Korean disordered speech database, incorporation of new parameters with cepstral coefficients outperforms the case with only cepstral coefficients.

• 전기의세계
• /
• 제13권3호
• /
• pp.28-37
• /
• 1964

It is the intention of this thesis to discriminate and investigate the cause of the shaft vibration of the vertical type hydroelectric power generator with respect to electrical, mechanical and hydraulic aspects, and to analyze the vibration which will occure by the each cause investigated above. In order to test the shaft vibration of No.1 generator in Hwachon, Korea new measurement method and measuring equipments were designed. In practice the shaft vibration of the generator was measured by above equipments and analyzed by the discriminative method. Detailed explanation for the designed measurement method and instruments is presented, and the results which I had tested three times for the generator No.1 in Hwachon power plant are added. As a appendix the mechanism and causes of the thrust bearing's wear and remarks for the runner are written.