• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.11
• /
• pp.44-52
• /
• 1996

This paper proposes a KLT-CVQ scheme using PCNN to improbe the quality of the reconstructed images at a given bit rate. By using the PCNN and classified vector quantization, we exploit the high energy compaction and compelte decorrelation capbilities of the KLT, and the pdf (probability density function) shape and space-filling advantages of the vQ to improve the performance of the proposed hybrid coding technique. In order to preserve the preceptual fetures such as the edge components in the reconstructed images, we classified the input image blocks according to the texture energy measures of the local statistics and vector-coded them adaptively, and thereby reduces the possible edge degradation in the reconstructed images. The results of the computer simulations show that the performance of the proposed KLT-CVQ is higher than that of the KLT-CSQ or the DCT-CVQ in the quality of the reconstructed images at a given bit rate.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.1E
• /
• pp.65-72
• /
• 1995

The effects of random errors in array weight and sensor positions on the performance of a Linearly constrained linear sensor array is analyzed in a weight vector space. It is observed that a nonorthogonality exists between an optimum weight vector and the steering vector of an interference direction du e to random errors. A novel approach to improving the nulling performance by compensating for the nonorthogonality is proposed. Computer simulation results are presented.

• ETRI Journal
• /
• v.39 no.4
• /
• pp.443-454
• /
• 2017

In this paper, we propose a fine-grained mobile application clustering model using retrofitted document embedding. To automatically determine the clusters and their numbers with no predefined categories, the proposed model initializes the clusters based on title keywords and then merges similar clusters. For improved clustering performance, the proposed model distinguishes between an accurate clustering step with titles and an expansive clustering step with descriptions. During the accurate clustering step, an automatically tagged set is constructed as a result. This set is utilized to learn a high-performance document vector. During the expansive clustering step, more applications are then classified using this document vector. Experimental results showed that the purity of the proposed model increased by 0.19, and the entropy decreased by 1.18, compared with the K-means algorithm. In addition, the mean average precision improved by more than 0.09 in a comparison with a support vector machine classifier.

• Proceedings of the KIEE Conference
• /
• 1991.11a
• /
• pp.140-143
• /
• 1991

This paper describes a compensation method for the rotor resistance variation of induction machines in speed sensor-less vector control system using MRAS(model reference adaptive system). In case of rotor resistance variation, the analysis of the conventional speed sensor-less vector control system using MRAS is presented and the compensation method for rotor resistance variation using Fuzzy logic is proposed. In order to confirm the performance of the proposed algorithm, computer simulation is performed.

• Proceedings of the KIEE Conference
• /
• 1997.07f
• /
• pp.2139-2142
• /
• 1997

The most important thing in vector control scheme is the knowledge of accurate electrical motor parameters. These parameters can computed by conventional motor test, such as no-load and locked rotor tests. However, the values from these tests are different from actual motor parameters, and the adjustment process of the parameters is time consuming. This paper presents an auto-tuning method for vector control of induction motor. The tuning algorithm is based on the rotor flux behavior of the induction motor for stepwise torque current command. The transient terminal voltage caused by the undesirable variation of the rotor flux is used for tuning the slip gain $K_5$ defined as the inverse of the rotor time constant. The electrical parameters of induction motor can also calculated by this method. The presented method is evaluated through the computer simulations.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.11 no.3
• /
• pp.204-210
• /
• 2011

This paper proposes a fuzzy support vector machine ($FSVM_n$) pattern classifier to classify the arrhythmia patterns of an electrocardiograph (ECG). The $FSVM_n$ is a pattern classifier which combines n-dimensional fuzzy membership functions with a slack variable of SVM. To evaluate the performance of the proposed classifier, the MIT/BIH ECG database, which is a standard database for evaluating arrhythmia detection, was used. The pattern classification experiment showed that, when classifying ECG into four patterns - NSR, VT, VF, and NSR, VT, and VF classification rate resulted in 99.42%, 99.00%, and 99.79%, respectively. As a result, the $FSVM_n$ shows better pattern classification performance than the existing SVM and FSVM algorithms.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.2
• /
• pp.549-557
• /
• 2016

In this paper we propose multiclassification method for large data sets by ensembling least squares support vector machines (LS-SVM) with principal components instead of raw input vector. We use the revised one-vs-all method for multiclassification, which is one of voting scheme based on combining several binary classifications. The revised one-vs-all method is performed by using the hat matrix of LS-SVM ensemble, which is obtained by ensembling LS-SVMs trained using each random sample from the whole large training data. The leave-one-out cross validation (CV) function is used for the optimal values of hyper-parameters which affect the performance of multiclass LS-SVM ensemble. We present the generalized cross validation function to reduce computational burden of leave-one-out CV functions. Experimental results from real data sets are then obtained to illustrate the performance of the proposed multiclass LS-SVM ensemble.

• Proceedings of the KIEE Conference
• /
• 1991.07a
• /
• pp.637-640
• /
• 1991

In this paper deals with linearized control of induction motor by vector control. Output equation induced from d-q axies voltage and current equation of induction moter. The condition of induced equation is that rotor's current of axies has 0 and state current of D axies which was driven by synchronous speed is constant. The fully digital controlled induction motor drive system based on the proposed linearized method and the control circuit of system consists of 16bits micro computer and all the function are implemented with software. When the voltage source inverter control with PI controller is empolyed, in spite of secondary resistance Rr Variation, the Vector control condition is satisfied.

• Bulletin of the Korean Mathematical Society
• /
• v.47 no.3
• /
• pp.551-561
• /
• 2010

In this paper we give a new characterization of real hypersurfaces of type B, that is, a tube over a totally geodesic $\mathbb{Q}P^n$ in complex two-plane Grassmannians $G_2(\mathbb{C}^{m+2})$, where m = 2n, with the Reeb vector $\xi$ belonging to the distribution $\mathfrak{D}$, where $\mathfrak{D}$ denotes a subdistribution in the tangent space $T_xM$ such that $T_xM$ = $\mathfrak{D}{\bigoplus}\mathfrak{D}^{\bot}$ for any point $x{\in}M$ and $\mathfrak{D}^{\bot}=Span{\xi_1,\;\xi_2,\;\xi_3}$.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.7
• /
• pp.2254-2263
• /
• 2000

This paper proposes an efficient predictive image-compression technique based on vector quantization of blocks of pels. In the proposed method edge contents of blocks control the selection of predictors and block shapes as well. The maximum number of bits assigned to quantizers has been in creased to 3bits/pel from 1/5bits/pel, the setting employed by forerunners in predictive vector quantization of images. This increase prevents the saturation in SNR observed in their results in high bit rates. The variable block shape is instrumental in eh reconstruction of edges. The adaptive procedure is controlled by means of he standard deviation ofp rediction errors generated by a default predictor; the standard deviation address a decision table which can be set up beforehand. eh proposed method is characterized by overall improvements in image quality over A-VQ-PE and A-DCT VQ, both of which are known for their efficient use of vector quantizers.