• Journal of KIISE:Software and Applications
• /
• v.29 no.3
• /
• pp.167-179
• /
• 2002

In this paper, we propose a new acoustic model for characterizing segmental features and an algorithm based upon a general framework of hidden Markov models (HMMs) in order to compensate the weakness of HMM assumptions. The segmental features are represented as a trajectory of observed vector sequences by a polynomial regression function because the single frame feature cannot represent the temporal dynamics of speech signals effectively. To apply the segmental features to pattern classification, we adopted segmental HMM(SHMM) which is known as the effective method to represent the trend of speech signals. SHMM separates observation probability of the given state into extra- and intra-segmental variations that show the long-term and short-term variabilities, respectively. To consider the segmental characteristics in acoustic model, we present segmental-feature HMM(SFHMM) by modifying the SHMM. The SFHMM therefore represents the external- and internal-variation as the observation probability of the trajectory in a given state and trajectory estimation error for the given segment, respectively. We conducted several experiments on the TIMIT database to establish the effectiveness of the proposed method and the characteristics of the segmental features. From the experimental results, we conclude that the proposed method is valuable, if its number of parameters is greater than that of conventional HMM, in the flexible and informative feature representation and the performance improvement.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.12
• /
• pp.1150-1158
• /
• 2010

In this study, the Polynomial-based Radial Basis Function Neural Networks is proposed as one of the recognition part of overall face recognition system that consists of two parts such as the preprocessing part and recognition part. The design methodology and procedure of the proposed pRBFNNs are presented to obtain the solution to high-dimensional pattern recognition problem. First, in preprocessing part, we use a CCD camera to obtain a picture frame in real-time. By using histogram equalization method, we can partially enhance the distorted image influenced by natural as well as artificial illumination. We use an AdaBoost algorithm proposed by Viola and Jones, which is exploited for the detection of facial image area between face and non-facial image area. As the feature extraction algorithm, PCA method is used. In this study, the PCA method, which is a feature extraction algorithm, is used to carry out the dimension reduction of facial image area formed by high-dimensional information. Secondly, we use pRBFNNs to identify the ID by recognizing unique pattern of each person. The proposed pRBFNNs architecture consists of three functional modules such as the condition part, the conclusion part, and the inference part as fuzzy rules formed in 'If-then' format. In the condition part of fuzzy rules, input space is partitioned with Fuzzy C-Means clustering. In the conclusion part of rules, the connection weight of pRBFNNs is represented as three kinds of polynomials such as constant, linear, and quadratic. Coefficients of connection weight identified with back-propagation using gradient descent method. The output of pRBFNNs model is obtained by fuzzy inference method in the inference part of fuzzy rules. The essential design parameters (including learning rate, momentum coefficient and fuzzification coefficient) of the networks are optimized by means of the Particle Swarm Optimization. The proposed pRBFNNs are applied to real-time face recognition system and then demonstrated from the viewpoint of output performance and recognition rate.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.3 s.309
• /
• pp.25-33
• /
• 2006

This paper proposed device-independent color matching algorithm based on the 3D RGB lookup table (LUT) between mobile camera and mobile LCD (Liquid Crystal Display) to improve the color-fidelity. Proposed algorithm is composed of thee steps, which is device characterization, gamut mapping, 3D RGB-LUT design. First, the characterization of mobile LCD is executed using the sigmoidal function, different from conventional method such as GOG (Gain Offset Gamma) and S-curve modeling, based on the observation of electro-optical transfer function of mobile LCD. Next, mobile camera characterization is conducted by fitting the digital value of GretagColor chart captured under the daylight environment (D65) and tristimulus values (CIELAB) using the polynomial regression. However, the CIELAB values estimated by polynomial regression exceed the maximum boundary of the CIELAB color space. Therefore, these values are corrected by linear compression of the lightness and chroma. Finally, gamut mapping is used to overcome the gamut difference between mobile camera and moible LCD. To implement the real-time processing, 3D RGB-LUT is designed based on the 3D RGB-LUT and its performance is evaluated and compared with conventional method.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.3
• /
• pp.35-44
• /
• 2000

In this paper, we propose a new trajectory model for characterizing segmental features and their interaction based upon a general framework of hidden Markov models. Each segment, a sequence of vectors, is represented by a trajectory of observed sequences. This trajectory is obtained by applying a new design matrix which includes transitional information on contiguous frames, and is characterized as a polynomial regression function. To apply the trajectory to the segmental HMM, the frame features are replaced with the trajectory of a given segment. We also propose the likelihood of a given segment and the estimation of trajectory parameters. The obervation probability of a given segment is represented as the relation between the segment likelihood and the estimation error of the trajectories. The estimation error of a trajectory is considered as the weight of the likelihood of a given segment in a state. This weight represents the probability of how well the corresponding trajectory characterize the segment. The proposed model can be regarded as a generalization of a conventional HMM and a parametric trajectory model. The experimental results are reported on the TIMIT corpus and performance is show to improve significantly over that of the conventional HMM.

• Membrane Journal
• /
• v.22 no.1
• /
• pp.23-34
• /
• 2012

In this study, we investigated the permeation property of single gases ($N_2$, $O_2$, $SF_6$, $CF_4$ through hollow fiber polymeric membrane (PSF, PC, PI) as a function of pressure and temperature to decide operating condition for $SF_6$ gas separation process. The results showed the gas permeation varied differentlydepending on the properties of gases and membrane. When permeance of each gases was represented as a function of temperature and pressure in 3 dimensional space, the surface of permeance was shown approximately flat. Thus, we established permeance models with forms of first-and second-order polynomial. These two models showed high goodness of fit. This indicates that the two polynomial models have enough applicability to predict the gas separation process.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.20 no.3
• /
• pp.311-319
• /
• 2007

A Meshfree is a method used to establish algebraic equations of system for the whole problem domain without the use of a predefined mesh for the domain discretization. A point interpolation method is based on combining radial and polynomial basis functions. Involvement of radial basis functions overcomes possible singularity Furthermore, the interpolation function passes through all scattered points in an influence domain and thus shape functions are of delta function property. This makes the implementation of essential boundary conditions much easier than the meshfree methods based on the moving least-squares approximation. This study aims to investigate a stress analysis of structural element between a meshfree method and the finite element method. Examples on cantilever type plate, hollow cylinder and stress concentration problems show that the accuracy and convergence rate of the meshfree methods are high.

• Geomechanics and Engineering
• /
• v.3 no.1
• /
• pp.61-81
• /
• 2011

In this paper, a complement to the Hoek-Brown criterion is proposed in order to derive the strength of anisotropic rock from strength of the corresponding truly intact rock. The complement is a decay function, which unlike other modifications or suggestions made in the past, is multiplied to the function of the original Hoek-Brown failure criterion for intact rock. This results in a combined and extended form of the criterion which describes the strength of anisotropic rock as a varying fraction of the corresponding truly intact rock strength. Statistical procedures and in particular regression analyses were conducted into data obtained in experiments conducted in the current research program and those collected from the literature in order to define the Hoek-Brown's criterion complement. The complement function was best described by a simple polynomial including only three constants to be empirically evaluated. Further investigations also showed that these constants can be related to the other readily available parameters of rock material which further facilitate determining the constants. A great and prime advantage of the proposed complement is that it is mathematically simple including the least possible number of empirical constants which are easily estimated with minimum experimental effort. Moreover, proposed concept does not suggests any change to the original Hoek-Brown criterion itself or its constants and serves whenever anisotropy does exist in the rock. This further implies on the possibility of using any other failure criterion for intact rock in conjunction with the compliment to reach the strength of anisotropic rock.

• Journal of the Microelectronics and Packaging Society
• /
• v.24 no.1
• /
• pp.67-73
• /
• 2017

An analysis of thermal degradation of epoxy based adhesive performed by thermogravimetry tests are presented in this study. Six different heating rates were employed for the weight change measurements. Based on the data, an Arrhenius type modeling equation was developed by calculating activation energies and proportional constants, and $n^{th}$ polynomial function was adopted to predict the weight change rates. The prediction results by the modeling was compared with the data using the average activation energy. It was found that the activation energy at the each heating rate was not same due to the different degradation kinetics, especially at the high heating rate. To overcome this pitfall, a new approach using exponential function series was introduced and employed. The calculation results showed very good agreements with the test data regardless of the heating rates.

• Journal of Korean Association for Spatial Structures
• /
• v.12 no.2
• /
• pp.45-53
• /
• 2012

This paper derives Isogeometric analysis and post-processing method of surface that are generated by NURBS basis function for accurate geometric modeling and structure analysis of free-form. By deforming these parameters that are consisted of control points, knots, polynomial, variable geometric models are derived. The basis function that is used to Isogeometric analysis is same to the basis function of NURBS that is used to generate geometric models. For performing isogeometric analysis, h-p-k refinement is performed without changing of original geometry. To visualize the results of isogeometric analysis that control points' displacements, post-processing method that is the interface method between IGES format and Rhinoceros is derived.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.6
• /
• pp.586-591
• /
• 2014

In this study, we introduce Radial Basis Function Neural Networks(RBFNNs) classifier using Artificial Bee Colony(ABC) algorithm in order to classify between precipitation event and non-precipitation event from given radar data. Input information data is rebuilt up through feature analysis of meteorological radar data used in Korea Meteorological Administration. In the condition phase of the proposed classifier, the values of fitness are obtained by using Fuzzy C-Mean clustering method, and the coefficients of polynomial function used in the conclusion phase are estimated by least square method. In the aggregation phase, the final output is obtained by using fuzzy inference method. The performance results of the proposed classifier are compared and analyzed by considering both QC(Quality control) data and CZ(corrected reflectivity) data being used in Korea Meteorological Administration.