• The Journal of the Acoustical Society of Korea
• /
• v.28 no.3
• /
• pp.268-272
• /
• 2009

In this paper, we propose a Gaussian Mixture Model (GMM) based incremental robust adaptation with a forgetting factor for the speaker verification. Speaker recognition system uses a speaker model adaptation method with small amounts of data in order to obtain a good performance. However, a conventional adaptation method has vulnerable to the outlier from the irregular utterance variations and the presence noise, which results in inaccurate speaker model. As time goes by, a rate in which new data are adapted to a model is reduced. The proposed algorithm uses an incremental robust adaptation in order to reduce effect of outlier and use forgetting factor in order to maintain adaptive rate of new data on GMM based speaker model. The incremental robust adaptation uses a method which registers small amount of data in a speaker recognition model and adapts a model to new data to be tested. Experimental results from the data set gathered over seven months show that the proposed algorithm is robust against outliers and maintains adaptive rate of new data.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.6 no.2
• /
• pp.206-211
• /
• 2011

In this paper, a fire detection algorithm based on video processing is proposed. At the first stage, background image extracted from CCTV video input signal, and then foreground image were separated by differencing CCTV input signal from background image. At the second stage, candidated area were extracted by using color information from foreground image. At the final stage, smoke or flame characteristic area were separated by using Gaussian mixture modeling applied to candidated area, and then fire can be detected. Through real experiments at the inner room, it is shown that the proposed system works well.

• Nuclear Engineering and Technology
• /
• v.26 no.4
• /
• pp.562-569
• /
• 1994

In this study improvement of transient analysis model, KOTRAC, for the passive reactor has been performed. In the KOTRAC, mixture drift flux model is adopted to simulate thermal hydraulic behavior, which can simulate ECCS injection in the passive plant. However, there is a difficulty to handle complete phase separation phenomena due to the near-zero density, which may occur in the pressurizer surge line or horizontal flow paths. In this study, a couple of model changes to over-come Courant limit feilure has been examined. One of key features is to substitute flow distribution parameters with Ishii's correlation. Corrected results are nil compared to those of RELAP/MOD3 analysis.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.29 no.4
• /
• pp.195-201
• /
• 2017

In this study, the viscosity of a $CO_2-gas$ mixture was investigated for the transportation of the captured $CO_2-gas$ in pipelines and for the designing of a thermal system, both of which involve the utilization of the $CO_2-gas$ mixture. The viscosities of the $CO_2-gas$ mixture, $CO_2+CH_4$, $CO_2+H_2S$, and $CO_2+N_2$ were predicted using three different models as follows : Chung, TRAPP, and REFPROP. The predictability values of the models were validated by comparing the estimated results with the experiment data for the $CO_2+CH_4$ and $CO_2+N_2$ under high-density conditions. The Chung model showed 2.41%, which is the lowest mean deviation of the prediction among the model. Based on the Chung model, the mixture mole fractions were changed from 0.9, 0.95, and 0.97, the mixture pressure was ranged from 80 bar to 120 bar by 10 bar, and the mixture temperature was varied from 310 K to 400 K by 10 K to observe the effects of the parameters on the mixture viscosity. Considering the high mole fraction of the $CO_2$ in the mixture, a significant variation of the mixture viscosity was observed close to the pseudo-critical temperature, and the viscosity for the $CO_2+H_2S$ mixture shows the highest values compared with those of the $CO_2+CH_4$ and $CO_2+N_2$.

• Journal of Distribution Research
• /
• v.12 no.4
• /
• pp.1-25
• /
• 2007

The purpose of this study is to provide the strategic implication of the Korean open air market by examining the factors affecting customer loyalty for various market segments as their competitive environment becomes more turbulent. We have undertaken empirical research that uses the methodology of a mixture regression modeling, as a way to ascertain the determinants of customer loyalty toward the Korean open air market, which should form the base of strategy for each segment. We construct a mixture regression model which uses perceived the Korean open air market value dimensions as explanatory variables, an income as a covariate variable, and a customer loyalty as a dependent variable. The analysis of results show that customers are statistically divided into four segments: 'Accessibility'(33.7%), 'Price'(29.7%), 'Shopping environment,'(22.0%), and 'Merchandising,'(14.5%) groups. The findings also showed that parameter estimates are different for each group, which indicates that the sensitivity to changes in the Korean traditional market perceived value and the income variable affecting customer loyalty vary among segments.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.6
• /
• pp.49-56
• /
• 1999

In many continuous speech recognition systems based on HMMs, decision tree-based state tying has been used for not only improving the robustness and accuracy of context dependent acoustic modeling but also synthesizing unseen models. To construct the phonetic decision tree, standard method performs one-level pruning using just single Gaussian triphone models. In this paper, two novel approaches, two-level decision tree and multi-mixture decision tree, are proposed to get better performance through more accurate acoustic modeling. Two-level decision tree performs two level pruning for the state tying and the mixture weight tying. Using the second level, the tied states can have different mixture weights based on the similarities in their phonetic contexts. In the second approach, phonetic decision tree continues to be updated with training sequence, mixture splitting and re-estimation. Multi-mixture Gaussian as well as single Gaussian models are used to construct the multi-mixture decision tree. Continuous speech recognition experiment using these approaches on BN-96 and WSJ5k data showed a reduction in word error rate comparing to the standard decision tree based system given similar number of tied states.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.9
• /
• pp.265-273
• /
• 2018

In this paper, we propose an acoustic event detection method in tunnels using non-negative tensor factorization (NTF) and hidden Markov model (HMM) applied to multi-channel audio signals. Incidents in tunnel are inherent to the system and occur unavoidably with known probability. Incidents can easily happen minor accidents and extend right through to major disaster. Most incident detection systems deploy visual incident detection (VID) systems that often cause false alarms due to various constraints such as night obstacles and a limit of viewing angle. To this end, the proposed method first tries to separate and detect every acoustic event, which is assumed to be an in-tunnel incident, from noisy acoustic signals by using an NTF technique. Then, maximum likelihood estimation using Gaussian mixture model (GMM)-HMMs is carried out to verify whether or not each detected event is an actual incident. Performance evaluation shows that the proposed method operates in real time and achieves high detection accuracy under simulated tunnel conditions.

• Journal of computational fluids engineering
• /
• v.17 no.1
• /
• pp.94-100
• /
• 2012

In this paper, the cavitating flows around a hydrofoil have been numerically investigated by using a 2-d multi-phase RANS flow solver based on pseudo-compressibility and a homogeneous mixture model on unstructured meshes. For this purpose, a vertex-centered finite-volume method was utilized in conjunction with 2nd-order Roe's FDS to discretize the inviscid fluxes. The viscous fluxes were computed based on central differencing. The Spalart-Allmaras one equation model was employed for the closure of turbulence. A dual-time stepping method and the Gauss-Seidel iteration were used for unsteady time integration. The phase change rate between the liquid and vapor phases was determined by Merkle's cavitation model based on the difference between local and vapor pressure. Steady state calculations were made for the modified NACA66 hydrofoil at several flow conditions. Good agreements were obtained between the present results and the experiment for the pressure coefficient on a hydrofoil surface. Additional calculation was made for cloud cavitation around the hydrofoil. The observation of the vapor structure, such as cavity size and shape, was made, and the flow characteristics around the cavity were analyzed. Good agreements were obtained between the present results and the experiment for the frequency and the Strouhal number of cavity oscillation.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.12 no.1
• /
• pp.36-43
• /
• 1988

A two phase model for the simulation of flame propagation has been developed and applied to a mixture of coal air. The effects associated with changes in the initial coal partial equivalence ratio and the initial diameter of particles on the structure of laminar flame propagation have been studied qualitatively and quantitatively. Especially the flame structure, the burning velocity, and the thermal behavior were evaluated. It was found that the radiative heat transfer absolutely dominates over the conduction mode. The increase in particle size was seen to contribute to an obvious increase in burning velocity for fuel lean and stoichiometric mixture. But for fuel rich mixture, the burning velocity was found to exhibit a weaker dependence on particle size.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.7
• /
• pp.847-851
• /
• 2004

In the original mixture density network(MDN), which was introduced by Bishop and Nabney, the parameters of the conditional probability density function are represented by the output vector of a single multi-layer perceptron. Among the recent modification of the MDNs, there is the so-called modified mixture density network, in which each of the priors, conditional means, and covariances is represented via an independent multi-layer perceptron. In this paper, we consider a further simplification of the modified MDN, in which the conditional means are linear with respect to the input variable together with the development of the MATLAB program for the simplification. In this paper, we first briefly review the original mixture density network, then we also review the modified mixture density network in which independent multi-layer perceptrons play an important role in the learning for the parameters of the conditional probability, and finally present a further modification so that the conditional means are linear in the input. The applicability of the presented method is shown via an illustrative simulation example.