• Proceedings of the Korean Statistical Society Conference
• /
• 2003.05a
• /
• pp.243-248
• /
• 2003

Fattorini(1986)의 통계량은 Shapiro와 Wilk의 일변량 정규분포를 위한 검정통계량을 다변량으로 확장한 것이다. 본 논문에서는 Kim과 Bickel(2003)에서 제안한 이변량 정규분포를 위한 검정통계량을 Fattorini(1986)의 방법을 이용하여 이변량 이상인 경우에도 실제적으로 사용가능하도록 일반화하였다. 제안된 통계량은 Fattorini(1986) 통계량의 근사통계량으로 생각할 수 있으며 표본의 크기가 클 때도 사용가능하다.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2018.11a
• /
• pp.202-203
• /
• 2018

각 학교(전문대 이상)에서는 국제물류 전문인력 양성을 위한 전공이 개설되어 관련 과목 교육을 하고 있다. 학제별로 개설된 전공과목들을 파악하고 전체 커리큘럼을 분석하여, 정규교육과정에서 물류 전문인력을 양성하는데 한계점에 대하여 제시하였다. 또한, 한계점을 바탕으로 앞으로 정규교육이 우수한 국제물류 인력의 양성을 위하여 어떤 방향 점으로 개선되어야 하는지 분석하였다.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2003.06a
• /
• pp.126-129
• /
• 2003

자동 음성 인식(ASR) 시스템에, 끝점 추출과 에너지 정규화는 중요한 역할을 하게 된다 그러나 낮은 SNR이나 nonstationary 환경에서, 기존 방법은 끝점 추출과 에너지 정규화에 있어서 자주 실패하게 되며, ASR을 급격히 열화시키곤 한다. ASR을 수행하기 위해, 최적의 필터에 3상태 천이도를 사용하고, 필터는 정확성과 강인함을 확실히 하기 위해 여러 이론들을 이용하여 설계하였고 여러 가지 잡음이 있는 음성 신호환경에서 거의 일정한 응답을 주었다. 검출된 끝점은 곧바로 에너지 정규화에 적용된다. 실험 결과는 제안된 알고리즘이 낮은 SNR에서 에러율을 크게 감소시키고 있다는 것을 보여준다.

• Phonetics and Speech Sciences
• /
• v.12 no.3
• /
• pp.65-71
• /
• 2020

Feature normalization is a method to reduce the effect of environmental mismatch between the training and test conditions through the normalization of statistical characteristics of acoustic feature parameters. It demonstrates excellent performance improvement in the traditional Gaussian mixture model-hidden Markov model (GMM-HMM)-based speech recognition system. However, in a deep neural network (DNN)-based speech recognition system, minimizing the effects of environmental mismatch does not necessarily lead to the best performance improvement. In this paper, we attribute the cause of this phenomenon to information loss due to excessive feature normalization. We investigate whether there is a feature normalization method that maximizes the speech recognition performance by properly reducing the impact of environmental mismatch, while preserving useful information for training acoustic models. To this end, we introduce the mean and exponentiated variance normalization (MEVN), which is a compromise between the mean normalization (MN) and the mean and variance normalization (MVN), and compare the performance of DNN-based speech recognition system in noisy and reverberant environments according to the degree of variance normalization. Experimental results reveal that a slight performance improvement is obtained with the MEVN over the MN and the MVN, depending on the degree of variance normalization.

• Journal of Cadastre & Land InformatiX
• /
• v.48 no.2
• /
• pp.99-108
• /
• 2018

The purpose of this study is to determine the precise height in mountainous areas of South Korea and Jiri mountain area was selected as a test bed for the study. Gravity observation and GNSS surveying were performed for 44 BM(Benchmark) points in the test bed and calculate the height and the height correction. In the calculation, the dynamic correction amount, the orthometric correction amount and the normal correction amount were calculated, and the dynamic height and orthometric height and the normal height were calculated considering each correction amount. The results showed that the difference between normal gravity and observed gravity and also the difference between orthometric correction and the normal correction. In addition, the results of the comparison of the present official BM height and the computed orthometric height in this study show that Korean height system should be shifted from the normal orthometric height system to the orthometric height system. Because the difference between the orthometric correction and the normal correction within the test bed indicated a distribution of at a minimum of -234.41 mm up to 196.925 mm, and the difference between the present official BM height and the calculated orthometric height were distributed from -0.121m to 0.011 m.

• Journal of Korea Water Resources Association
• /
• v.56 no.12
• /
• pp.871-881
• /
• 2023

This study investigates the geomorphic changes and Bed Relief Index of the river downstream of the Yeongju Dam by Nays2DH, a two-dimensional numerical model, in order to grasp the dynamics of the downstream river while applying various flow patterns such as pulse discharge. It shows that the geomorphic and the bed elevations changes are the largest under the condition of the normalized pulse discharge. The total change in the riverbed is 29.88 m for uniform flow, 27.46 m for normalized hydrograph, 29.63 m for pulse flow and 31.87 m for pulse flow with normalized hydrograph which result in the largest variation in scour and deposition. The Bed Relief Index (BRI) increases with time under conditions of uniform flow, pulse flow and pulse flow with normalized hydrograph. However, BRI increased rapidly until 30 hrs after the peak flow (14 hrs), but decreased from 56 hrs under the condition of normalized hydrograph. Therefore, the condition of normalized hydrograph gives greater dynamics than the condition of a single flood or constant flow, and the dynamics increase downstream than upstream, resulting in an effect on improving the environment of the river downstream of the dam.

• 대학교육
• /
• s.100
• /
• pp.83-87
• /
• 1999

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.30 no.4
• /
• pp.353-362
• /
• 2012

3D building modeling from airborne Lidar without model regularization may cause positional errors or topological inconsistency in building models. Regularization of 3D building models, on the other hand, restricts the types of models which can be reconstructed. To resolve these issues, this paper modelled 3D buildings from airborne Lidar by building model regularization which considers more various types of buildings. Building points are first segmented into roof planes by clustering in feature space and segmentation in object space. Then, 3D building models are reconstructed by consecutive adjustment of planes, lines, and points to satisfy parallelism, symmetry, and consistency between model components. The experimental results demonstrated that the method could make more various types of 3d building models with regularity. The effects of regularization on the positional accuracies of models were also analyzed quantitatively.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.17 no.4
• /
• pp.221-231
• /
• 2005

The present study is deeply concerned with the reliability design method(Level III) for caisson breakwaters using expected sliding distance, and the objectives of this study are to propose the employment of a doubly-truncated normal distribution and to present the validity for it. In this study, therefore, the explanations are made for consideration of effects of uncertain factors, and a clear basis that the doubly-truncated normal distribution should be employed in the computation process of expected sliding distance by Monte-Carlo simulation is presented with introduction of the employment method. Even though only caisson breakwaters are treated in this paper, the employment of doubly-truncated normal distribution can be applied to various coastal structures as well as other engineering fields, and therefore it is expected that the present study will be extended in various fields.

• The Journal of the Acoustical Society of Korea
• /
• v.34 no.4
• /
• pp.316-320
• /
• 2015

In this paper, the pole filtering concept is applied to the Mel-frequency cepstral coefficient (MFCC) feature vectors in the conventional cepstral mean normalization (CMN) and cepstral mean and variance normalization (CMVN) frameworks. Additionally, performance of the cepstral mean and scale normalization (CMSN), which uses scale normalization instead of variance normalization, is evaluated in speech recognition experiments in noisy environments. Because CMN and CMVN are usually performed on a per-utterance basis, in case of short utterance, they have a problem that reliable estimation of the mean and variance is not guaranteed. However, by applying the pole filtering and scale normalization techniques to the feature normalization process, this problem can be relieved. Experimental results using Aurora 2 database (DB) show that feature normalization method combining the pole-filtering and scale normalization yields the best improvements.