• The Korean Journal of Applied Statistics
• /
• v.22 no.4
• /
• pp.759-770
• /
• 2009

A problem of separating signals from noises is considered, when they are randomly mixed in the observation. It is assumed that the noise follows a Gaussian distribution and the signal follows a Gamma distribution, thus the underlying distribution of an observation will be a mixture of Gaussian and Gamma distributions. The parameters of the mixture model will be estimated from the EM algorithm. Then the signals and noises will be classified by a fixed threshold approach based on multiple testing using positive false discovery rate and Bayes error. The proposed method is applied to a real optical emission spectroscopy data for the quantitative analysis of inclusions. A simulation is carried out to compare the performance with the existing method using 3 sigma rule.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.11a
• /
• pp.259-262
• /
• 2009

본 논문에서는 정확한 비디오 객체 분할을 위한 글로벌 가우시안 모델 기반의 이동 외부 그림자영역 검출방법을 제안한다. 이 방법은 현재 픽셀과 배경 픽셀의 컬러 벡터간의 사이 각을 가중치 함수로 변환하고, 이를 그림자 모델의 확률 밀도에 곱하여 구한 값을 그림자 검출에 사용하고 이를 다시 그림자 모델의 입력으로 하여 검출된 픽셀 들의 분포가 자동으로 영상의 실제 그림자 분포에 근접하게 하였다. 또한, 잘못 검출된 그림자 영역을 제거하기 위해 영역의 위치 정보를 이용한다. 실험 결과를 통해 제안하는 방법은 적응적으로 그림자를 검출하면서도 높은 분할 정확도를 가지고 있음을 보인다.

• Proceedings of the IEEK Conference
• /
• 2009.05a
• /
• pp.381-383
• /
• 2009

본 논문에서는 대표적인 시스템 모델링 도구중의 하나인 RBF 뉴럴 네트워크(Radial Basis Function Neural Network)를 설계하고 모델을 최적화하기 위하여 최적화 알고리즘인 PSO(Particle Swarm Optimization) 알고리즘을 이용하였다. 즉, 모델의 최적화에 주요한 영향을 미치는 모델의 파라미터들을 PSO 알고리즘을 이용하여 동정한다. 제안된 RBF 뉴럴 네트워크는 은닉층에서의 활성함수로서 일반적으로 많이 사용되어지는 가우시안 커널함수를 사용한다. 더 나아가 모델의 최적화를 위하여 각 커널함수의 중심값은 HCM 클러스터링에 기반을 두어 중심값을 결정하고, PSO 알고리즘을 통하여 가우시안 커널함수의 분포상수, 은닉층에서의 노드 수 그리고 다수의 입력을 가질 경우 입력의 종류를 동정한다. 제안한 모델의 성능을 평가하기 위해 Mackey-Glass 시계열 공정 데이터를 적용하였으며 제안된 모델의 근사화와 일반화 능력을 분석한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.544-546
• /
• 2002

본 논문은 CDHMM(Continuous Density Hidden Markov Model)의 훈련하는 방법을 동적 다중 그룹 혼합 가중치(Dynamic Mutli-Group mixture weight)을 이용하여 재구성하는 방법을 제안한다. 음성은 Hidden 상태열에 의하여 특성화되고, 각 상태는 가중된 혼합 가우시안 밑도 함수에 의해 표현된다. 음성신호를 더욱더 정확하게 계산하려면 각 상태를 위한 가우시안 함수를 더욱더 많이 사용해야 하며 이것은 많은 계산량이 요구된다. 이러한 문제는 가우시안 분포 확률의 통계적인 평균을 이용하면 계산량을 줄일 수 있다. 그러나 이러한 기존의 방법들은 다양한 화자의 발화속도와 가중치의 적용이 적합하지 못하여 인식률을 저하시키는 단점을 가지고 있다. 이 문제를 다양한 화자의 발화속도에 적합하도록 화자의 화자의 발화속도에 따라 동적으로 5개의 그룹으로 구성하고 동적 다중 그룹 혼합 가중치를 적용하여 CDHMM 파라미터를 재구성함으로써 8.5%의 인식율이 증가되었다.

• Journal of Korea Multimedia Society
• /
• v.14 no.1
• /
• pp.85-93
• /
• 2011

Previous existing vocabulary recognition programs calculate general vector values from a database, so they can not process phonemes that form during a search. And because they can not create a model for phoneme data, the accuracy of the Gaussian model can not secure. Therefore, in this paper, we recommend use of the Bhattacharyya distance measurement method based on the features of the phoneme-thus allowing us to improve the recognition rate by picking up accurate phonemes and minimizing recognition of similar and erroneous phonemes. We test the Gaussian model optimization through share continuous probability distribution, and we confirm the heighten recognition rate. The Bhattacharyya distance measurement method suggest in this paper reflect an average 1.9% improvement in performance compare to previous methods, and it has average 2.9% improvement based on reliability in recognition rate.

• Journal of KIISE:Software and Applications
• /
• v.35 no.7
• /
• pp.417-423
• /
• 2008

In this paper we present a probabilistic method for source separation in the case here each source has a certain temporal structure. We tackle the problem of source separation by maximum pseudo-likelihood estimation, representing the latent function which characterizes the temporal structure of each source by a random process with a Gaussian prior. The resulting pseudo-likelihood of the data is Gaussian, determined by a mixing matrix as well as by the predictive mean and covariance matrix that can easily be computed by Gaussian process (GP) regression. Gradient-based optimization is applied to estimate the demixing matrix through maximizing the log-pseudo-likelihood of the data. umerical experiments confirm the useful behavior of our method, compared to existing source separation methods.

• Journal of Digital Convergence
• /
• v.10 no.7
• /
• pp.167-172
• /
• 2012

In vocabulary recognition using HMM(Hidden Markov Model) by model for the observation of a discrete probability distribution indicates the advantages of low computational complexity, but relatively low recognition rate has the disadvantage that require sophisticated smoothing process. Gaussian mixtures in order to improve them with a continuous probability density CHMM (Continuous Hidden Markov Model) model is proposed for the optimization of the library system. In this paper is system configuration thread control in recognition Gaussian mixtures model provides a model to optimize of the CHMM vocabulary recognition. The result of applying the proposed system, the recognition rate of 98.1% in vocabulary recognition, respectively.

• Journal of Korea Multimedia Society
• /
• v.12 no.8
• /
• pp.1155-1163
• /
• 2009

This paper proposes an automatic facial expression generation system of vector graphic character using gaussian process model. Proposed method extracts the main feature vectors from twenty-six facial data of character redefined based on Russell's internal emotion state. Also by using new gaussian process model, SGPLVM, we find low-dimensional feature data from extracted high-dimensional feature vectors, and learn probability distribution function (PDF). All parameters of PDF are estimated by maximization the likelihood of learned expression data, and these are used to select wanted facial expressions on two-dimensional space in real time. As a result of simulation, we confirm that proposed facial expression generation tool is working in the small facial expression datasets and can generate various facial expressions without prior knowledge about relation between facial expression and emotion.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.6
• /
• pp.675-681
• /
• 2019

Noise removal is a pre-requisite procedure in image processing, and various methods have been studied depending on the type of noise and the environment of the image. However, for image processing with high-frequency components, conventional additive white Gaussian noise (AWGN) removal techniques are rather lacking in performance because of the blurring phenomenon induced thereby. In this paper, we propose an algorithm to minimize the blurring in AWGN removal processes. The proposed algorithm sets the high-frequency and the low-frequency component filters, respectively, depending on the pixel properties in the mask, consequently calculating the output of each filter with the addition or subtraction of the input image to the reference. The final output image is obtained by adding the weighted data calculated using the standard deviations and the Gaussian distribution with the output of the two filters. The proposed algorithm shows improved AWGN removal performance compared to the existing method, which was verified by simulation.

• Journal of Digital Convergence
• /
• v.16 no.11
• /
• pp.335-340
• /
• 2018

Commercialized speech recognition systems that have an accuracy recognition rates are used a learning model from a type of speaker dependent isolated data. However, it has a problem that shows a decrease in the speech recognition performance according to the quantity of data in noise environments. In this paper, we proposed the vector quantization based speech recognition performance improvement using maximum log likelihood in Gaussian distribution. The proposed method is the best learning model configuration method for increasing the accuracy of speech recognition for similar speech using the vector quantization and Maximum Log Likelihood with speech characteristic extraction method. It is used a method of extracting a speech feature based on the hidden markov model. It can improve the accuracy of inaccurate speech model for speech models been produced at the existing system with the use of the proposed system may constitute a robust model for speech recognition. The proposed method shows the improved recognition accuracy in a speech recognition system.