• The Journal of the Acoustical Society of Korea
• /
• v.24 no.4
• /
• pp.210-216
• /
• 2005

We Propose a new feature normalization scheme based on eigenspace for achieving robust speech recognition. In general, mean and variance normalization (MVN) is Performed in cepstral domain. However, another MVN approach using eigenspace was recently introduced. in that the eigenspace normalization Procedure Performs normalization in a single eigenspace. This Procedure consists of linear PCA matrix feature transformation followed by mean and variance normalization of the transformed cepstral feature. In this method. 39 dimensional feature distribution is represented using only a single eigenspace. However it is observed to be insufficient to represent all data distribution using only a sin91e eigenvector. For more specific representation. we apply unique na independent eigenspaces to cepstra, delta and delta-delta cepstra respectively in this Paper. We also normalize training data in eigenspace and get the model from the normalized training data. Finally. a feature space rotation procedure is introduced to reduce the mismatch of training and test data distribution in noisy condition. As a result, we obtained a substantial recognition improvement over the basic eigenspace normalization.

• The Korean Journal of Zoology
• /
• v.34 no.2
• /
• pp.217-227
• /
• 1991

Ultrastmctura changes of multivesicular bodies and mitochondria of oocytes of PseudopotamU- Ia occelata Moore were examined with transmission electron microscope in order to follow the process of yolk formation. Yolk granules begin to form at the previtellogenic stages of 50 $\mu$m diameter from multivesicular bodies. Small vesicles and membranous structures within the multivesicular bodies are fused to form the precursors of core bodies of yolk granules. Some vesicles from cytoplasm are also coalesced into the multivesicular bodies. Mature yolk granules are composed of electron-dense core bodies which are seperated from each other by electronopaque small vesicle-like structures. Structural changes of cristae into vesicular shapes and increase in electron density of matrix in mitochondria strongly suggested that mitochondria are in the process of transformation. The transformed mitochondria appear to be basic structures which later become multivesicular bodies.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.2
• /
• pp.237-244
• /
• 2020

This paper describes an implement of Simultaneous Localization and Mapping(SLAM) in two dimensional space. The method uses Invariant Extended Kalman Filter(IEKF), which transforms the state variables and measurement variables so that the transformed variables constitute a linear space when variables called the invariant quantities are kept constant. Therefore, the IEKF guarantees convergence provided in the invariant quantities are kept constant. The proposed IEKF approach uses Lie group matrix for the transformation. The method is tested through simulation, and the results show that the Kalman gain is constant as it is the case for the linear Kalman filter. The coherence between the estimated locations of the vehicle and the detected objects verifies the estimation performance of the method.

• Journal of Ocean Engineering and Technology
• /
• v.27 no.2
• /
• pp.75-79
• /
• 2013

This study was carried out to investigate the precipitation behavior of the ${\chi}$ phase and the impact toughness of 25%Cr-7%Ni-2%Mo-4%W super duplex stainless steel. The ${\chi}$ phase was precipitated in the early stage of aging, and with the increasing aging time. However, after reaching a maximum value, the number decreased as a result of the gradual transformation of the ${\chi}$ phase into the ${\sigma}$-phase. It was proved that the ${\chi}$ phase was an intermetallic compound, which represented a lower nickel concentration, higher chromium and molybdenum concentrations, and very higher tungsten concentration compared to the matrix phases. It also showed higher molybdenum and tungsten concentrations than the ${\sigma}$ phase. The decomposition of the ferrite phase into the ${\gamma}_2$ and ${\sigma}$ phases was retarded by W substitution for Mo. Thus, the number of ${\chi}$ phases increased. The impact value was decreased by the substitution of W for Mo. The impact toughness rapidly decreased with time when the ${\chi}$ phase began to precipitate in the initial stage of aging. The impact toughness was, therefore, greatly influenced for the precipitation of the ${\chi}$ phase.

• Journal of the Korean Ceramic Society
• /
• v.38 no.5
• /
• pp.405-411
• /
• 2001

Compacts of a mixture of fine $\alpha$-Si$_3$N$_4$powders, 6% $Y_2$O$_3$and 1% $Al_2$O$_3$were attrition milled time on phase and microstructural development in silicon nitride ceramics. The sintered surface and the interior showed different behaviors in phase and microstructral developments. Increased oxygen content with increased milling time of powder mixture leads to the formation of Si$_2$$N_2$O phase at temperatures as low as 155$0^{\circ}C$. Si$_2$$N_2$O is stable in the interior of the samples but unstable in the surface region of the specimen sintered at higher temperature. This results in a duplex structure where the interior consists of Si$_2$$N_2$O grains dispersed in $\beta$-Si$_3$N$_4$matrix and a surface which contains only $\beta$-Si$_3$N$_4$. The alpha to beta phase transformation and the microstructural development are shown to be influenced by the formation and decomposition of the Si$_2$$N_2$O.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.1
• /
• pp.6-11
• /
• 2011

In this paper, we propose a gait-based human identification system using eigenfeature regularization and extraction (ERE). First, a gait feature for human identification which is called gait energy image (GEI) is generated from walking sequences acquired from a camera sensor. In training phase, regularized transformation matrix is obtained by applying ERE to the gallery GEI dataset, and the gallery GEI dataset is projected onto the eigenspace to obtain galley features. In testing phase, the probe GEI dataset is projected onto the eigenspace created in training phase and determine the identity by using a nearest neighbor classifier. Experiments are carried out on the CASIA gait dataset A to evaluate the performance of the proposed system. Experimental results show that the proposed system is better than previous works in terms of correct classification rate.

• Smart Structures and Systems
• /
• v.26 no.3
• /
• pp.373-390
• /
• 2020

Hybrid simulation (HS) is a versatile tool for structural performance evaluation under dynamic loads. Although real structural responses are often multiple-directional owing to an eccentric mass/stiffness of the structure and/or excitations not along structural major axes, few HS in this field takes into account structural responses in multiple directions. Multi-directional loading is more challenging than uni-directional loading as there is a nonlinear transformation between actuator and specimen coordinate systems, increasing the difficulty of suppressing loading error. Moreover, redundant actuators may exist in multi-directional hybrid simulations of large-scale structures, which requires the loading strategy to contain ineffective loading of multiple actuators. To address these issues, lately a new strategy was conceived for accurate reproduction of desired displacements in bi-directional hybrid simulations (BHS), which is characterized in two features, i.e., iterative displacement command updating based on the Jacobian matrix considering nonlinear geometric relationships, and force-based control for compensating ineffective forces of redundant actuators. This paper performs performance validation and application of this new mixed loading strategy. In particular, virtual BHS considering linear and nonlinear specimen models, and the diversity of actuator properties were carried out. A validation test was implemented with a steel frame specimen. A real application of this strategy to BHS on a full-scale 2-story frame specimen was performed. Studies showed that this strategy exhibited excellent tracking performance for the measured displacements of the control point and remarkable compensation for ineffective forces of the redundant actuator. This strategy was demonstrated to be capable of accurately and effectively reproducing the desired displacements in large-scale BHS.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• v.29 no.1
• /
• pp.10-23
• /
• 2007

Hallmarks of clinical behaviors of adenoid cystic carcinoma(ACC) of salivary glands are the delayed onset of vascular metastasis and poor responses to classical chemotherapeutic agents. Poor prognoses from salivary ACC are caused by lung metastases that are resistant to conventional therapy. Therefore, cellular and molecular characteristics that influence the dissemination of metastatic cells are important for the design of more effective treatment of salivary ACC. Tumor angiogenesis has been known to be essential for the distant metastasis of malignant cells. So, we determined expressions of angiogenic proteins in benign (pleomorphic adenoma) and malignant (ACC, mucoepidermoid carcinoma) tumors of salivary glands and compared each other and to those in oral squamous cell carcinoma. Using surgical specimens, we performed immunohistochemical assays with anti-vascular endothelial growth factor (VEGF), VEGF receptor-2 (VEGFR-2), phosphorylated VEGFR-2 (pVEGFR-2), matrix metalloproteinase (MMP)-9, and interleukin (IL)-8 antibodies. Most angiogenic factors were overexpressed in malignant salivary tumors than in pleomorphic adenoma which is benign nature. Moreover, ACC demonstrated more expression of VEGFR-2 than that of squamous cell carcinoma which used as control. Conclusively, these data show those angiogenic factors produced by salivary gland tumors may affect the propagation and metastasis of malignant cells of salivary tumors, and could be used as biomarkers for the malignant transformation of salivary gland tumors. Prospectively, although further studies will be needed, these biomarkers related to angiogenesis can be molecular targets for the therapy of salivary ACC, which has propensity for delayed vascular metastasis.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.2
• /
• pp.50-55
• /
• 2002

The standard approach of image resampling is to fit the original image with continuous model and resample the function at a desired rate. We used the B-spline function as the continuous model because it oscillates less than the others. The main purpose of this paper is the derivation of a nonuniform optimal resampling algorithm. To derive it, needing approximation can be computed in three steps: 1) determining the I-spline coefficients by matrix inverse process, 2) obtaining the transformed-spline coefficients by the optimal resampling algorithm derived from the orthogonal projection theorem, 3) converting of the result back into the signal domain by indirect B-spline transformation. With these methods, we can use B-spline in the non-uniform resampling, which is proved to be a good kernel in uniform resampling, and can also verify the applicability from our experiments.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.4
• /
• pp.271-277
• /
• 2018

Generating an epipolar image which is removed a y-parallax from an original image is an essential technique for creating a 3D stereoscopic model or producing a map. In epipolar image production, there is a method of generating epipolar images by estimating the relative orientation parameters after matching the extracted distinct points in two images and a method of generating epipolar images by using the baseline and rotation angles of the two images after determining the exterior orientation parameters In this study, it was proposed a methodology to generate epipolar images using direction cosine in the exterior orientation parameters of the input images, and a method to use the transformation matrix for easy calculation when converting from the original image to the epipolar image. The applicability of the proposed methodology was evaluated by using images taken from the fixed wing and rotary wing drones. As a result, it was found that epipolar images were generated regardless of the type of drones.