• Speech Sciences
• /
• v.15 no.2
• /
• pp.55-65
• /
• 2008

Recently, the multi-model based speech recognizer has been used quite successfully for noisy speech recognition. For the selection of the reference HMM (hidden Markov model) which best matches the noise type and SNR (signal to noise ratio) of the input testing speech, the estimation of the SNR value using the VAD (voice activity detection) algorithm and the classification of the noise type based on the GMM (Gaussian mixture model) have been done separately in the multi-model framework. As the SNR estimation process is vulnerable to errors, we propose an efficient method which can classify simultaneously the SNR values and noise types. The KL (Kullback-Leibler) distance between the single Gaussian distributions for the noise signal during the training and testing is utilized for the classification. The recognition experiments have been done on the Aurora 2 database showing the usefulness of the model compensation method in the multi-model based speech recognizer. We could also see that further performance improvement was achievable by combining the probability density function of the MCT (multi-condition training) with that of the reference HMM compensated by the D-JA (data-driven Jacobian adaptation) in the multi-model based speech recognizer.

• Journal of KSNVE
• /
• v.10 no.5
• /
• pp.769-774
• /
• 2000

The purpose of the work is to present successful applications of a modified wavelet shrinkage method for the accurate and fast estimation of a transfer function. Although the experimental process of determining a transfer function introduces not only Gaussian but also non-Gaussian noises, most existing estimation methods are based only on a Gaussian noise model. To overcome this limitation, we propose to employ a modified wavelet shrinkage method in which L1 -based median filtering and L2 -based wavelet shrinkage are applied repeatedly. The underlying theory behind this approach is briefly explained and the superior performance of this modified wavelet shrinkage technique is demonstrated by a numerical example.

• 한국정보컨버전스학회:학술대회논문집
• /
• 2008.06a
• /
• pp.113-116
• /
• 2008

In this paper we suggest two novel methods for an implementation of the spot detection phase in the 2-DE gel image analysis program. The one is the adaptive thresholding method for eliminating noises and the other is the asymmetric diffusion model for spot matching. Remained noises after the preprocessing phase cause the over-segmentation problem by the next segmentation phase. To identify and exclude the over-segmented background regions, il we use a fixed thresholding method that is choosing an intensity value for the threshold, the spots that are invisible by one's human eyes but mean very small amount proteins which have important role in the biological samples could be eliminated. Accordingly we suggest the adaptive thresholding method which comes from an idea that is got on statistical analysis for the prominences of the peaks. There are the Gaussian model and the diffusion model for the spot shape model. The diffusion model is the closer to the real spot shapes than the Gaussian model, but spots have very various and irregular shapes and especially asymmetric formation in x-coordinate and y-coordinate. The reason for irregularity of spot shape is that spots could not be diffused perfectly across gel medium because of the characteristics of 2-DE process. Accordingly we suggest the asymmetric diffusion model for modeling spot shapes. In this paper we present a brief explanation ol the two methods and experimental results.

• Proceedings of the Korea Electromagnetic Engineering Society Conference
• /
• 2000.11a
• /
• pp.16-19
• /
• 2000

We propose a deterministic vector channel simulation model satisfying not only rigorous temporal correlation but also arbitrary spatial correlation using the method of Doppler phase difference sampling. The model is more efficient than the conventional PN filtered Gaussian model with coloring process in evaluating the laboratory performance of mobile communication systems employing adaptive way antennas or space diversity.

• Smart Structures and Systems
• /
• v.24 no.2
• /
• pp.243-256
• /
• 2019

Determination of the most meaningful structural modes and gaining insight into how these modes evolve are important issues for long-term structural health monitoring of the long-span bridges. To address this issue, modal parameters identified throughout the life of the bridge need to be compared and linked with each other, which is the process of mode tracking. The modal frequencies for a long-span bridge are typically closely-spaced, sensitive to the environment (e.g., temperature, wind, traffic, etc.), which makes the automated tracking of modal parameters a difficult process, often requiring human intervention. Machine learning methods are well-suited for uncovering complex underlying relationships between processes and thus have the potential to realize accurate and automated modal tracking. In this study, Gaussian mixture model (GMM), a popular unsupervised machine learning method, is employed to automatically determine and update baseline modal properties from the identified unlabeled modal parameters. On this foundation, a new mode tracking method is proposed for automated mode tracking for long-span bridges. Firstly, a numerical example for a three-degree-of-freedom system is employed to validate the feasibility of using GMM to automatically determine the baseline modal properties. Subsequently, the field monitoring data of a long-span bridge are utilized to illustrate the practical usage of GMM for automated determination of the baseline list. Finally, the continuously monitoring bridge acceleration data during strong typhoon events are employed to validate the reliability of proposed method in tracking the changing modal parameters. Results show that the proposed method can automatically track the modal parameters in disastrous scenarios and provide valuable references for condition assessment of the bridge structure.

• ETRI Journal
• /
• v.41 no.4
• /
• pp.437-451
• /
• 2019

Outlier detection techniques play an important role in enhancing the reliability of data communication in wireless sensor networks (WSNs). Considering the importance of outlier detection in WSNs, many outlier detection techniques have been proposed. Unfortunately, most of these techniques still have some potential limitations, that is, (a) high rate of false positives, (b) high time complexity, and (c) failure to detect outliers online. Moreover, these approaches mainly focus on either temporal outliers or spatial outliers. Therefore, this paper aims to introduce novel algorithms that successfully detect both temporal outliers and spatial outliers. Our contributions are twofold: (i) modifying the Hampel Identifier (HI) algorithm to achieve high accuracy identification rate in temporal outlier detection, (ii) combining the Gaussian process (GP) model and graph-based outlier detection technique to improve the performance of the algorithm in spatial outlier detection. The results demonstrate that our techniques outperform the state-of-the-art methods in terms of accuracy and work well with various data types.

• Structural Engineering and Mechanics
• /
• v.78 no.2
• /
• pp.175-186
• /
• 2021

Due to the increasing environmental pollution caused by scrap tires, a solution is being sought to recycle and use them in a field of civil engineering, i.e., construction. This paper will provide a brief overview of previous researches that give detailed information on the advantages and disadvantages, considering the microstructural and mechanical characteristics of self-compacting concrete, when waste tire rubber as an aggregate is added. With this aim, a database of 144 different mixtures of self-compacting concrete with partial substitute of natural aggregate with recycled tire rubber (self-compacting rubberized concrete, SCRC) provided by various researchers was created. In this study we show that Gaussian process regression (GPR) modelling is an appropriate method for predicting compressive strength of SCC with recycled tire rubber particles and is in accordance with the results displayed by SEM images.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.1
• /
• pp.48.1-48.1
• /
• 2021

Gaussian process regression (GPR) is a powerful method used for model-independent analysis of cosmological observations. In GPR, it is important to decide an input mean function and hyperparameters that affect the reconstruction results. Depending on how the input mean function and hyperparameters are determined in the literature, I divide into four main applications for GPR and compare their results. In particular, a zero mean function is commonly used as an input mean function, which may be inappropriate for reconstructing cosmological observations such as the distance modulus. Using mock data based on Pantheon compilation of type Ia supernovae, I will point out the problem of using a zero input and suggest a new way to deal with the input mean function.

• Convergence Security Journal
• /
• v.8 no.4
• /
• pp.161-166
• /
• 2008

This paper proposes a new approach to align 3D data sets by using curvatures of feature surface. We use the Gaussian curvatures and the covariance matrix which imply the physical characteristics of the model to achieve registration of unaligned 3D data sets. First, the physical characteristics of local area are obtained by the Gaussian curvature. And the camera position of 3D range finder system is calculated from by using the projection matrix between 3D data set and 2D image. Then, the physical characteristics of whole area are obtained by the covariance matrix of the model. The corresponding points can be found in the overlapping region with the cross-projection method and it concentrates by removed points of self-occlusion. By the repeatedly the process discussed above, we finally find corrected points of overlapping region and get the optimized registration result.

• Journal of the Korean Statistical Society
• /
• v.33 no.4
• /
• pp.353-365
• /
• 2004

In this paper, we define the weighted U-empirical process for simple linear model and show the weak convergence to a Gaussian process under some conditions. Then we illustrate the usage of our result with examples. In the appendix, we derive the variance of the weighted U-empirical distribution function.