• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.366-369
• /
• 2005

Background subtraction is mainly used for the real-time extraction and tracking of moving objects from image sequences. In the outdoor environment, there are many changeable factor such as gradually changing illumination, swaying trees and suddenly moving objects, which are to be considered for the adaptive processing. Normally, GMM(Gaussian Mixture Model) is used to subtract the background adaptively considering the various changes in the scenes, and the adaptive GMMs improving the real-time performance were worked. This paper, for on-line background subtraction, applied the improved adaptive GMM, which uses the small constant for learning rate ${\alpha}$ and is not able to speedily adapt the suddenly movement of objects, So, this paper proposed and evaluated the dynamic control method of ${\alpha}$ using the adaptive selection of the number of component distributions and the global variances of pixel values.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.6
• /
• pp.167-176
• /
• 2005

This paper describes how the state tying model based on the decision tree which is one of Acoustic models used for speech recognition optimizes the model by reducing the number of mixture Gaussians of the output probability distribution. The state tying modeling uses a finite set of questions which is possible to include the phonological knowledge and the likelihood based decision criteria. And the recognition rate can be improved by increasing the number of mixture Gaussians of the output probability distribution. In this paper, we'll reduce the number of mixture Gaussians at the highest point of recognition rate by clustering the Gaussians. Bhattacharyya and Euclidean method will be used for the distance measure needed when clustering. And after calculating the mean and variance between the pair of lowest distance, the new Gaussians are created. The parameters for the new Gaussians are derived from the parameters of the Gaussians from which it is born. Experiments have been performed using the STOCKNAME (1,680) databases. And the test results show that the proposed method using Bhattacharyya distance measure maintains their recognition rate at $97.2\%$ and reduces the ratio of the number of mixture Gaussians by $1.0\%$. And the method using Euclidean distance measure shows that it maintains the recognition rate at $96.9\%$ and reduces the ratio of the number of mixture Gaussians by $1.0\%$. Then the methods can optimize the state tying model.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.05a
• /
• pp.407-409
• /
• 2015

In this paper, a hardware implementation of MOD(Moving Object Detection) algorithm is described, which is based GMM(Gaussian Mixture Model) and background subtraction. The EGML(Effective Gaussian Mixture Learning) is used to model and update background. Some approximations of EGML calculations are applied to reduce hardware complexity, and pipelining technique is used to improve operating speed. Gaussian parameters are adjustable according to various environment conditions to achieve better MOD performance. MOD processor is verified by using FPGA-in-the-loop verification, and it can operate with 109 MHz clock frequency on XC5VSX95T FPGA device.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.7
• /
• pp.519-527
• /
• 2003

Speech is much influenced by the existence of outliers which are introduced by such an unexpected happenings as additive background noise, change of speaker's utterance pattern and voice detection errors. These kinds of outliers may result in severe degradation of speaker recognition performance. In this paper, we proposed the GMM based on robust principal component analysis (RPCA-GMM) using M-estimation to solve the problems of both ouliers and high dimensionality of training feature vectors in speaker identification. Firstly, a new feature vector with reduced dimension is obtained by robust PCA obtained from M-estimation. The robust PCA transforms the original dimensional feature vector onto the reduced dimensional linear subspace that is spanned by the leading eigenvectors of the covariance matrix of feature vector. Secondly, the GMM with diagonal covariance matrix is obtained from these transformed feature vectors. We peformed speaker identification experiments to show the effectiveness of the proposed method. We compared the proposed method (RPCA-GMM) with transformed feature vectors to the PCA and the conventional GMM with diagonal matrix. Whenever the portion of outliers increases by every 2%, the proposed method maintains almost same speaker identification rate with 0.03% of little degradation, while the conventional GMM and the PCA shows much degradation of that by 0.65% and 0.55%, respectively This means that our method is more robust to the existence of outlier.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.6 s.38
• /
• pp.187-194
• /
• 2005

The background subtraction method is mainly used for the real-time extraction and tracking of moving objects from image sequences. In the outdoor environment, there are many changeable environment factors such as gradually changing illumination, swaying trees and suddenly moving objects , which are to be considered for an adaptive processing. Normally, GMM(Gaussian Mixture Model) is used to subtract the background by considering adaptively the various changes in the scenes, and the adaptive GMMs improving the real-time Performance were Proposed and worked. This paper, for on-line background subtraction, employed the improved adaptive GMM, which uses the small constant for learning rate a and is not able to speedily adapt the suddenly movement of objects, So, this paper Proposed and evaluated the dynamic control method of a using the adaptive selection of the number of component distributions and the global variances of pixel values.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.3
• /
• pp.141-149
• /
• 2005

Tn general. Gaussian mixture model (GMM) is used to estimate the speaker model from the speech for speaker identification. The parameter estimates of the GMM are obtained by using the Expectation-Maximization (EM) algorithm for the maximum likelihood (ML) estimation. However the EM algorithm has such drawbacks that it depends heavily on the initialization and it needs the number of mixtures to be known. In this paper, to solve the above problems of the EM algorithm. we propose an EM algorithm with the initialization based on incremental ${\cal}k-means$ for GMM. The proposed method dynamically increases the number of mixtures one by one until finding the optimum number of mixtures. Whenever adding one mixture, we calculate the mutual relationship between it and one of other mixtures respectively. Finally. based on these mutual relationships. we can estimate the optimal number of mixtures which are statistically independent. The effectiveness of the proposed method is shown by the experiment for artificial data. Also. we performed the speaker identification by applying the proposed method comparing with other approaches.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.2
• /
• pp.127-134
• /
• 2010

In continuous vocabulary recognition system by probability distribution of clustering method has used model parameters of an advance estimate to generated each contexts for phoneme data surely needed but it has it's bad points of gaussian model the accuracy unsecure of composed model for phoneme data. To improve suggested probability distribution mixed gaussian model to optimized that phoneme data search supported configuration thread system. This paper of configuration thread system has used extension facet classification user phoneme configuration thread information offered gaussian model the accuracy secure. System performance as a result of represent vocabulary dependence recognition rate of 98.31%, vocabulary independence recognition rate of 97.63%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.5
• /
• pp.1047-1054
• /
• 2015

This paper presents an acoustic model transform method using untranscribed speech database for improved speech recognition. In the presented model transform method, an adapted GMM is obtained by employing the conventional adaptation method, and the most similar Gaussian component is selected from the adapted GMM. The bias vector between the mean vectors of the clean GMM and the adapted GMM is used for updating the mean vector of HMM. The presented GAMT combined with MAP or MLLR brings improved speech recognition performance in car noise and speech babble conditions, compared to singly-used MAP or MLLR respectively. The experimental results show that the presented model transform method effectively utilizes untranscribed speech database for acoustic model adaptation in order to increase speech recognition accuracy.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10b
• /
• pp.682-684
• /
• 2004

제안하는 방법은 분산량이 큰 객체에 대하여 여러 개의 가우시안을 이용하여 다양체를 분해하는 알고리즘이다. 제안하는 방법은 단순하지만 빠르게 다양체를 근사시키는 여러 개의 가우시안을 생성한다. 또한, 가우시안 혼합 모델과 유사하나 보다 빠른 연산시간을 보장하며 Outlier에 대한 신뢰성을 향상 시켜준다. 제안하는 알고리즘은 우리가 수집한 다 인종(동양인, 혹인, 백인, 히스패닉) 얼굴 데이터 베이스 QQVGA영상에서 100%의 검출률과 0개의 오분류의 높은 성능을 도출하였다

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.2
• /
• pp.395-406
• /
• 2017

The purpose of this study is to identify the pattern of daily electricity demand through clustering and classification. The hourly data was collected by KPS (Korea Power Exchange) between 2008 and 2012. The time trend was eliminated for conducting the pattern of daily electricity demand because electricity demand data is times series data. We have considered k-means clustering, Gaussian mixture model clustering, and functional clustering in order to find the optimal clustering method. The classification analysis was conducted to understand the relationship between external factors, day of the week, holiday, and weather. Data was divided into training data and test data. Training data consisted of external factors and clustered number between 2008 and 2011. Test data was daily data of external factors in 2012. Decision tree, random forest, Support vector machine, and Naive Bayes were used. As a result, Gaussian model based clustering and random forest showed the best prediction performance when the number of cluster was 8.