• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1984년도 추계학술발표회 논문집
• /
• pp.73-77
• /
• 1984

This paper reports on a Voiced-Unvoiced-Silence Classification of speech for Korean Speech Recognition. In this paper, it is describe a method which uses a Pattern Recognition Technique for classifying a given speech segment into the three classes. Best result is obtained with the combination using ZCR, P1, Ep and classification error rate is less than 1%.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1993년도 한국자동제어학술회의논문집(국내학술편); Seoul National University, Seoul; 20-22 Oct. 1993
• /
• pp.168-172
• /
• 1993

As the worst-case analysis for interval plants, a conjecture whether the supremum of the integral of square error(ISE) is attained at the extreme point such as vertices, Kharitonov vertices, CB segment, and edges is suggested. We present a sufficient condition for which the worst performance index occurs at one ofvertices of uncertain parameter space. Numerical examples are also given.

• 대한인간공학회지
• /
• 제22권1호
• /
• pp.31-42
• /
• 2003

This study describes a modification of the technique for human error analysis in nuclear power plants (NPPs) which adopts advanced Man-Machine Interface (MMI) features based on computerized working environment, such as LCOs. Flat Panels. Large Wall Board, and computerized procedures. Firstly, the state of the art on human error analysis methods and efforts were briefly reviewed. Human error analysis method applied to NPP design has been THERP and ASEP mainly utilizing Swain's HRA handbook, which has not been facilitated enough to put the varied characteristics of MMI into HRA process. The basic concepts on human errors and the system safety approach were revisited, and adopted the process of FMEA with the new definition of Error Segment (ESJ. A modified human error analysis process was suggested. Then, the suggested method was applied to the failure of manual pump actuation through LCD touch screen in loss of feed water event in order to verify the applicability of the proposed method in practices. The example showed that the method become more facilitated to consider the concerns of the introduction of advanced MMI devices, and to integrate human error analysis process not only into HRA/PRA but also into the MMI and interface design. Finally, the possible extensions and further efforts required to obtain the applicability of the suggested method were discussed.

• 대한전자공학회논문지SP
• /
• 제40권6호
• /
• pp.114-123
• /
• 2003

본 논문에서는 다각형 기반 윤곽선 부호화를 위한 새로운 정점 선택 기법을 제안한다. 객체의 모양을 효율적으로 표현하기 위해서 정점 선택 과정에서 기존의 최대 거리 척도뿐만 아니라 곡률 정보를 이용한다. 제안된 기법은 두 단계 과정으로 구성되어 있다. 첫 번째 단계에서는 곡률이 큰 윤곽선 화소들을 CSS (curvature scale space)를 이용하여 키 정점(key vortices)으로 선택한다. 이 점들은 객체의 윤곽선을 여러 개의 부분 윤곽선(contour segment)들로 나누며, 각 부분 윤곽선은 양 끝점이 인접한 두 개의 키 정점인 하나의 열린 윤곽선으로 독립적으로 처리된다. 두 번째 단계에서는 각각의 부분 윤곽선에서 주어진 허용 오차(D/sub max//sup */)를 만족하면서 최소 개수의 정점을 선택하기 위해 순차적 정점 선택 기법을 이용하여 정점을 선택한다. 그리고 선택된 정점들은 오차 영역 면에서 최적의 위치를 찾기 위해 동적 프로그래밍(DP : dynamic programming) 기법을 사용하여 조정된다. 제안된 기법과 기존의 기법들의 근사 성능을 비교하기 위해 실험 결과를 제시한다.

• 한국도로학회논문집
• /
• 제16권5호
• /
• pp.133-142
• /
• 2014

PURPOSES : The objective was to develop the advanced method which could not explain each observation's specific characteristic in the present negative binomial method that results in under-estimation of the standard error(t-value inflation) and affects the confidence of whole derived results. METHODS : This study dealt with traffic accidents occurring within interchange segment on highway main line with RPNB(Random Parameter Negative Binomial) method that enables to take account of heterogeneity. RESULTS : As a result, AADT and lighting installation type on the road were revealed to have random parameter and in terms of other geometric variables, all were derived as fixed parameter(same effect on every segment). Also, marginal effects were adapted to analyze the relative effects on traffic accidents. CONCLUSIONS : This study proves that RPNB method which considers each observation's specific characteristics is better fitted to the accident data with geometrics. Thus, it is recommended that RPNB model or other methods which could consider the heterogeneity needs to be adapted in accident analysis.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1997년도 추계학술대회
• /
• pp.459-463
• /
• 1997

In this paper, we present the elimination of ECG artifact from the polysomnographic EEG and EOG. The idea of this method is that the ECG synchronized EEG segment is detected from ECG and regard samples of that segment a missing signal. After this, we used two interpolation methods to recover the missing segment. One is the Lagrange Polynomial Interpolation Method and the other is the Least Square Error AR Interpolation method. We tested those methods by applying to simulated signals. AR methods works well enough to reject the artifact about 10% of the main artifact level. We practically applied to real EEG and EOG signals. We also developed the algorithm to detect whether the artifact level is high or not. If the artifact level is high, then the interpolations are applied.

• 대한전자공학회논문지
• /
• 제20권3호
• /
• pp.13-20
• /
• 1983

본 논문은 도형 패턴에 대한 모서리를 결정하는 방법을 제안하였다. 도형 패턴의 윤곽선에 대한 라인 시그먼트(line segment)의 연결 상태로부터 후보 링크를 설정한다. 그리고 라인 시그먼트의 연결점으로부터 설정된 후보 링크에 대한 편차를 계산하여 모서리의 위치를 결정하는 알고리즘과 출. 철 구조의 모서리를 구분하여 식별하는 알고리즘을 제안한다. 그 결과 Freeman의 방법과 비교하여 훨씬 유연하고 효과적임을 화인하였다.

• 한국음향학회지
• /
• 제10권5호
• /
• pp.69-76
• /
• 1991

본 논문에서는 통계적 방법에 의한 음소의 자동분할에 관한 알고리즘을 제안하였다. 우선 음성 신호를 AR 모델로 모델링한 후 스펙트럼이 변화하기 전과 변화한 후의 모델에 대해서 likelihood ratio 와 mutual information을 고려한 test statistics 로부터 모델 계수가 변화하는 곳을 예측해 내고 이 곳을 음소의 경계로 판단한다. 이 경우 검파되지 못하는 대부분의 음소는 짧은 자음이었으며 Signed front-to-back maximum area ratio을 이용하여 개선하였다. 또한 false alarm error을 줄이기 위해 두 segment 사이의 distortion 으로부터 smoothing을 하였다. 3명의 화자에 대한 실험 결과 non-detection error는 10%, false alarm error는 20% 정도로 나타났지만 화자간에 알고리즘의 성능 변화가 거의 없으 며 특히 분할된 경계치 분포는 전체 음소의 90% 이상이 이 30ms 이내에 위치하였다.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 2000년도 하계학술발표대회 논문집 제19권 1호
• /
• pp.119-122
• /
• 2000

In this paper, we implement an elementary system to correct accent, pronunciation, and intonation in English spoken by non-native English speakers. In case of the accent evaluation, energy and pitch information are used to find stressed syllables, and then we extract the segment information of input patterns using a dynamic time warping method to discriminate and evaluate accent position. For the pronunciation evaluation. we utilize the segment information using the same algorithm as in accent evaluation and calculate the spectral distance measure for each phoneme between input and reference. For the intonation evaluation. we propose nine pattern of slope to estimate pitch contour, then we grade test sentences by accumulated error obtained by the distance measure and estimated slope. Our result shows that 98 percent of accent and 71 percent of pronunciation evaluation agree with perceptual measure. As the result of the intonation evaluation. system represent the similar order of grade for the four sentences having different intonation patterns compared with perceptual evaluation.

• 전자공학회논문지B
• /
• 제33B권3호
• /
• pp.79-87
• /
• 1996

A new method for recognizing and locating partially occluded or overlapped two-dimensional objects regardless of their size, translation, and rotation, is presented. Dominant points approximating occuluding contoures of objects are generated by finding local maxima of smoothed k-cosine function, and then used to guide the contour segment matching procedure. Primitives between the dominant points are produced by projecting the local contours onto the line between the dominant points. Robust classification of primitives. Which is crucial for reliable partial shape matching, is performed using adaptive resonance theory (ART2). The matched primitives having similar scale factors and rotation angles are detected in the hough space to identify the presence of the given model in the object scene. Finally the translation vector is estimated by minimizing the mean squred error of the matched contur segment pairs. This model-based matching algorithm may be used in diveerse factory automation applications since models can be added or changed simply by training ART2 adaptively without modifying the matching algorithm.