• Communications for Statistical Applications and Methods
• /
• 제20권5호
• /
• pp.395-404
• /
• 2013

This paper studies the skewness of the absolute value GARCH(1, 1) models with Gaussian mixture innovations (Gaussian mixture AVGARCH(1, 1) models). The maximum estimated-likelihood estimator (MELE) employed (a two- step estimation method in order to estimate the skewness of Gaussian mixture AVGARCH(1, 1) models. Through the real data analysis, the adequacy of adopting Gaussian mixture innovations is exhibited in reflecting the skewness of two major Korean stock indices.

• Journal of Positioning, Navigation, and Timing
• /
• 제9권1호
• /
• pp.23-31
• /
• 2020

It is essential to estimate the vehicle localization for an autonomous safety driving. In particular, since LIDAR provides precise scan data, many studies carried out to estimate the vehicle localization using LIDAR and pre-generated map. The road marking always exists on the road because of provides driving information. Therefore, it is often used for map information. In this paper, we propose to generate the Gaussian mixture map based on road-marking information and localization method using this map. Generally, the probability distributions map stores the single Gaussian distribution for each grid. However, single resolution probability distributions map cannot express complex shapes when grid resolution is large. In addition, when grid resolution is small, map size is bigger and process time is longer. Therefore, it is difficult to apply the road marking. On the other hand, Gaussian mixture distribution can effectively express the road marking by several probability distributions. In this paper, we generate Gaussian mixture map and perform vehicle localization using Gaussian mixture map. Localization performance is analyzed through the experimental result.

• 한국전자통신학회논문지
• /
• 제7권4호
• /
• pp.733-739
• /
• 2012

본 논문에서는 시간과 기후에 따라 변화하는 영상의 밝기와 색상 변화에도 강건한 연기검출 알고리즘을 제안한다. 제안하는 연기검출 알고리즘은 입력영상과 배경영상의 차영상을 이용하여 후보영역을 설정하고, 후보영역 차영상의 Gaussian 혼합모델 특징 계수를 비교하여 연기를 판별한다. 시간과 기후에 대응하기 위하여 입력영상의 평균 밝기와 색상을 기준으로 후보영역 설정을 위한 임계값을 4 단계로 구분한다. 후보영역에 대한 차영상의 Gaussian 혼합모델의 밝기 평균값을 기준으로 클러스터를 정렬하고, 클러스터 간의 Gaussian 혼합모델 특징 계수를 비교하여 연기를 판별한다. 제안하는 알고리즘을 미디어전용 DSP로 구현하고 야외에 설치된 카메라의 영상에 대하여 연기검출 실험을 통하여 효율적으로 연기를 검출할 수 있음 보인다.

• ETRI Journal
• /
• 제43권1호
• /
• pp.74-81
• /
• 2021

The mixture model is a very powerful and flexible tool in clustering analysis. Based on the Dirichlet process and parsimonious Gaussian distribution, we propose a new nonparametric mixture framework for solving challenging clustering problems. Meanwhile, the inference of the model depends on the efficient online variational Bayesian approach, which enhances the information exchange between the whole and the part to a certain extent and applies to scalable datasets. The experiments on the scene database indicate that the novel clustering framework, when combined with a convolutional neural network for feature extraction, has meaningful advantages over other models.

• IEIE Transactions on Smart Processing and Computing
• /
• 제2권6호
• /
• pp.332-338
• /
• 2013

We propose a method for target detection in Infrared images. In order to effectively detect a target region from an image with noises and clutters, spatial information of the target is first considered by analyzing pixel distributions of projections in horizontal and vertical directions. These distributions are represented as Gaussian distributions, and Gaussian Mixture Model is created from these distributions in order to find thresholding points of the target region. Through analyzing the calculated Gaussian Mixture Model, the target region is detected by eliminating various backgrounds such as noises and clutters. This is performed by using a novel thresholding method which can effectively detect the target region. As experimental results, the proposed method has achieved better performance than existing methods.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 하계종합학술대회 논문집(4)
• /
• pp.195-198
• /
• 2002

We propose an efficient mixture Gaussian synthesis method for decision tree based state tying that produces better context-dependent models in a short period of training time. This method makes it possible to handle mixture Gaussian HMMs in decision tree based state tying algorithm, and provides higher recognition performance compared to the conventional HMM training procedure using decision tree based state tying on single Gaussian GMMs. This method also reduces the steps of HMM training procedure. We applied this method to training of PBS, and we expect to achieve a little point improvement in phoneme accuarcy and reduction in training time.

• Communications for Statistical Applications and Methods
• /
• 제25권6호
• /
• pp.633-645
• /
• 2018

Gaussian error distributions are a common choice in traditional regression models for the maximum likelihood (ML) method. However, this distributional assumption is often suspicious especially when the error distribution is skewed or has heavy tails. In both cases, the ML method under normality could break down or lose efficiency. In this paper, we consider the log-concave and Gaussian scale mixture distributions for error distributions. For the log-concave errors, we propose to use a smoothed maximum likelihood estimator for stable and faster computation. Based on this, we perform comparative simulation studies to see the performance of coefficient estimates under normal, Gaussian scale mixture, and log-concave errors. In addition, we also consider real data analysis using Stack loss plant data and Korean labor and income panel data.

• 한국음향학회지
• /
• 제29권1호
• /
• pp.56-61
• /
• 2010

본 논문에서는 HMM 기반의 연속음성인식에서 음향모델의 최적화 기법을 논한다. 대부분의 음성인식 시스템에서 HMM 상태별로 동일한 개수의 가우시안 성분 (mixture component)을 사용해 왔다. 그러나, 음향 모델링에 사용되는 데이터 샘플의 개수는 HMM상태별로 다르므로 이에 따른 최적화를 수행할 경우 모델 파라미터의 개수를 효과적으로 줄일 수 있을 뿐 아니라, 디코딩 단계에서 음성인식기의 속도 및 인식 성능 개선이 기대된다. 본 연구에서 제안한 방법은 기존에 알려진 MDL (minimum description length) 기반의 음향모델 최적화 방법에서 가우시안 성분들의 통합과정에 가우시안 성분의 가중치 정보 (mixture weight)를 반영하도록 개선하였다. 인식 실험 결과, 제안한 방법은 가우시안 성분의 가중치를 반영하지 않는 기존 방법에 비해 향상된 최적화 성능을 보임을 확인할 수 있었다.

• 전기전자학회논문지
• /
• 제23권1호
• /
• pp.302-305
• /
• 2019

Non-orthogonal multiple access (NOMA) has been considered for the fifth generation (5G) mobile networks to provide high system capacity and low latency. We calculate the channel capacity for the weak channel user in NOMA and the channel capacity region for NOMA. In this paper, Gaussian mixture channel is compared to the additive white Gaussian noise (AWGN) channel. Gaussian mixture channel is modeled when we assume the practical signal modulation for the inter user interference, such as the binary phase shift keying (BPSK) modulation. It is shown that the channel capacity with BPSK inter user interference is better than that with Gaussian inter user interference. We also show that the channel capacity region with BPSK inter user interference is larger than that with Gaussian inter user interference. As a result, NOMA could perform better in the practical environments.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제30권5_6호
• /
• pp.454-460
• /
• 2003

이 논문은 입술의 형태를 효과적으로 인식하는 방법을 제안하였다. 입술은 PDM(Point Distribution Model)을 기반으로 점들의 집합으로 표현하였다. 주성분 분석법을 적용하여 입술 모델을 구하고 모델에서 사용하는 형태계수의 분포를 GMM(Gaussian Mixture Model)을 이용하여 구하였다. 이 과정에서 계수를 정하기 위하여 EM(Expectation Maximization) 알고리듬을 사용하였다. 입술 경계선 모델은 입술을 구성하는 각 점과 주변 영역에서의 화소간 변화를 이용하여 구성하였으며 입술 탐색시 사용되었다. 여러 영상을 대상으로 실험한 결과 좋은 결과를 얻었다.