• Journal of KIISE:Software and Applications
• /
• v.29 no.8
• /
• pp.521-528
• /
• 2002

We propose a new approach to the feature selection for the classification of handwritten characters using two-dimensional(2D) wavelet packet bases. To extract key features of an image data, for the dimension reduction Principal Component Analysis(PCA) has been most frequently used. However PCA relies on the eigenvalue system, it is not only sensitive to outliers and perturbations, but has a tendency to select only global features. Since the important features for the image data are often characterized by local information such as edges and spikes, PCA does not provide good solutions to such problems. Also solving an eigenvalue system usually requires high cost in its computation. In this paper, the original data is transformed with 2D wavelet packet bases and the best discriminant basis is searched, from which relevant features are selected. In contrast to PCA solutions, the fast selection of detailed features as well as global features is possible by virtue of the good properties of wavelets. Experiment results on the recognition rates of PCA and our approach are compared to show the performance of the proposed method.

• Journal of the Korean Geosynthetics Society
• /
• v.7 no.2
• /
• pp.41-47
• /
• 2008

This paper describes the results of model experiments in the laboratory, which were conducted to assess the behavior characteristics of geosynthetic reinforced soil walls according to different surcharge loads and reinforcement types. The model walls were built in the box having dimension, 100 cm tall, 140 cm long, and 100cm wide. Three types of geosynthetics, geonet, geogrid A and geogrid B, are used as the reinforcements. Decomposed granite soil (SM) was used as a backfill material. Seven model walls are constructed and tested. After the construction of the model wall, the LVDTs are installed to obtain the displacements of the wall face. As the results of the model tests, the maximum horizontal displacements of the model walls occurred due to uniform surcharge pressure were measured at the 0.7H from the bottom of the wall. The more the reinforcement strength increases, the more the wall displacements decrease, and also the reduction ratio of the wall displacement decrease with increasing the surcharge pressure.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.39 no.7
• /
• pp.721-727
• /
• 2015

The mechanical components for wind turbines are mainly manufactured using open-die forging. This research introduces an advanced forging method to produce the door frame of the tubular wind turbine tower. The advantages of this new forging method are an increase in the raw material utilization ratio and a reduction in energy cost. In the conventional method, the door frame is hot forged with a hydraulic press and amounts of material are machined out because of the shape difference between the forged and final machine products. The proposed forging method is composed of hot forging and ring rolling processes to increase the material utilization ratio. The effectiveness of this new forging method is deeply related to the ring rolled blank dimension before the final forging. To get the optimal ring rolled blank, forged shape prediction using the finite element analysis method was applied. The forged dimensions produced by the new forging method were verified through the first article production.

• Journal of Digital Contents Society
• /
• v.16 no.4
• /
• pp.605-613
• /
• 2015

This paper addresses the problem to make sparse the projection matrix in pattern recognition method. Recently, the size of computer program is often restricted in embedded systems. It is very often that developed programs include some constant data. For example, many pattern recognition programs use the projection matrix for dimension reduction. To improve the recognition performance, very high dimensional feature vectors are often extracted. In this case, the projection matrix can be very big. Recently, RSR(roated sparse regression) method[1] was proposed. This method has been proved one of the best algorithm that obtains the sparse matrix. We propose three methods to improve the RSR; outlier removal, sampling and elastic net RSR(E-RSR) in which the penalty term in RSR optimization function is replaced by that of the elastic net regression. The experimental results show that the proposed methods are very effective and improve the sparsity rate dramatically without sacrificing the recognition rate compared to the original RSR method.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.24 no.1
• /
• pp.132-142
• /
• 2000

The objective of this study is to experimentally investigate the noise propagating characteristics, the noise reduction mechanism and the performance of a slotted tube attached at the exit plane of a circular convergent nozzle. The experiment is performed through the systematic change of the jet pressure ratio and the slot length under the condition of two kinds of open area ratios, 25% and 51%. The open area ratio calculated by the tube length equivalent for the slot length is defined as the ratio of the total slot area to the surface area of a slotted tube. The experimental results for the near and far field sound, the visualization of jet structures and the static pressure distributions in the jet passing through a slotted tube are presented and explained in comparison with those for a simple tube. The propagating characteristics of supersonic jet noises from the slotted tube is closely connected with the slot length rather than the open area ratio, and its propagating pattern is similar to the simple tube. It is shown that the slotted tube has a good performance to suppress the shock-associated noise as well as the turbulent mixing noise in the range of a limited jet pressure and slot dimension. The considerable suppression of the shock‘associated noise is mainly due to the pressure relief caused by the high-speed jets passing through the slots on the tube. Both the strength of shock waves and the interval between them in a jet plume are decreased by the pressure relief. Moreover, the pressure relief is divided into the gradual and the sudden relief depending upon the open area ratio of the slotted tube. Consequently, the shock waves in a jet plume are also changed by the type of pressure relief. The gradual pressure relief caused by the slotted tube with the open area ratio 25% generates the weak oblique shock waves. On the contrary, the weak normal shock waves appear due to the sudden pressure relief caused by the slotted tube with the open area ratio 51%.

• The korean journal of orthodontics
• /
• v.47 no.1
• /
• pp.11-20
• /
• 2017

Objective: This randomized controlled trial aimed to compare the stability of mandibular arch orthodontic treatment outcomes between passive self-ligating and conventional systems during 6 months of retention. Methods: Forty-seven orthodontic patients with mild to moderate crowding malocclusions not requiring extraction were recruited based on inclusion criteria. Patients (mean age $21.58{\pm}2.94years$) were randomized into two groups to receive either passive self-ligating ($Damon^{(R)}$ 3MX, n = 23) or conventional system (Gemini MBT, n = 24) orthodontic treatment. Direct measurements of the final sample comprising 20 study models per group were performed using a digital caliper at the debonding stage, and 1 month, 3 months, and 6 months after debonding. Paired t-test, independent t-test, and non-parametric test were used for statistical analysis. Results: A significant increase (p < 0.01) in incisor irregularity was observed in both self-ligating and conventional system groups. A significant reduction (p < 0.01) in second interpremolar width was observed in both groups. Mandibular arch length decreased significantly (p = 0.001) in the conventional system group but not in the self-ligating system group. A similar pattern of stability was observed for intercanine width, first interpremolar width, intermolar width, and arch depth throughout the 6-month retention period after debonding. Comparison of incisor irregularity and arch dimension changes between self-ligating system and conventional system groups during the 6 months were non-significant. Conclusions: The stability of treatment outcomes for mild to moderate crowding malocclusions was similar between the self-ligating system and conventional system during the first 6 months of retention.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.133-138
• /
• 2005

Automotive pulleys play a key role in driving the cooling pump, oil pump, air-conditioner and so on by using an engine power. Researches on design processes and technologies of the pulleys can be found in many literatures. On the other hand, the areas related to manufacturing processes of the pulleys have been treated negligently. Vast data extracted from various information systems are transformed, integrated, and summarized to become a special database for helping users make a decision. The database, namely the data warehouse has been popularly used in the marketing and customer management of enterprises and recently applied to improve the design and manufacturing processes. In this study the manufacturing process of pulleys were analyzed through the intensive investigation of shop-floors and the interviews with workers and managers. The defects generated during a manufacturing process were categorized in a few types and the causes of defects examined for extracting the dominant parameters in the setup process for producing pulleys. As the first step to construct the data warehouse for the manufacturing processes of pulleys, authors proposed its architecture focused on the reduction of defect rate during the setup process.

• The KIPS Transactions:PartB
• /
• v.12B no.6 s.102
• /
• pp.715-720
• /
• 2005

This paper presents a hybrid recognition method of first moment of face image and principal component analysis(PCA). First moment is applied to reduce the dimension by shifting to the centroid of image, which is to exclude the needless backgrounds in the face recognitions. PCA is implemented by single layer neural network which has a teaming rule of Foldiak algorithm. It has been used as an alternative method for numerical PCA. PCA is to derive an orthonormal basis which directly leads to dimensionality reduction and possibly to feature extraction of face image. The proposed method has been applied to the problems for recognizing the 48 face images(12 Persons $\ast$ 4 scenes) of 64$\ast$64 pixels. The 3 distances such as city-block, Euclidean, negative angle are used as measures when match the probe images to the nearest gallery images. The experimental results show that the proposed method has a superior recognition performances(speed, rate). The negative angle has been relatively achieved more an accurate similarity than city-block or Euclidean.

• The KIPS Transactions:PartB
• /
• v.11B no.4
• /
• pp.485-490
• /
• 2004

In data representation, the clustering performs a grouping process which combines given data into some similar clusters. The various similarity measures have been used in many researches. But, the validity of clustering results is subjective and ambiguous, because of difficulty and shortage about objective criterion of clustering. The fuzzy clustering provides a good method for subjective clustering problems. It performs clustering through the similarity matrix which has fuzzy membership value for assigning each object. In this paper, for objective fuzzy clustering, the clustering algorithm which joins principal components analysis as a dimension reduction model with bayesian learning as a statistical learning theory. For performance evaluation of proposed algorithm, Iris and Glass identification data from UCI Machine Learning repository are used. The experimental results shows a happy outcome of proposed model.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.29 no.7
• /
• pp.508-514
• /
• 2018

An electroencephalogram (EEG) evaluates the electrical activity generated by brain cell interactions that occur during brain activity, and an EEG can evaluate the brain activity caused by hand movement. In this study, a 16-channel EEG was used to measure the EEG generated before and after hand movement. The measured data can be classified as a supervised learning model, a support vector machine (SVM). To shorten the learning time of the SVM, a feature extraction and vector dimension reduction by filtering is proposed that minimizes motion-related information loss and compresses EEG information. The classification results showed an average of 72.7% accuracy between the sitting position and the hand movement at the electrodes of the frontal lobe.