• Journal of Korea Multimedia Society
• /
• v.9 no.11
• /
• pp.1435-1444
• /
• 2006

We present an effective and real-time face detection method based on Principal Component Analysis(PCA) and Support Vector Machines(SVMs). We extract simple Haar-like features from training images that consist of face and non-face images, reinterpret the features with PCA, and select useful ones from the large number of extracted features. With the selected features, we construct a face detector using an SVM appropriate for binary classification. The face detector is not affected by the size of a training data set in a significant way, so that it showed 90.1 % detection rates with a small quantity of training data. it can process 8 frames per second for $320{\times}240$ pixel images. This is an acceptable processing time for a real-time system.

• Biomedical Science Letters
• /
• v.25 no.1
• /
• pp.99-106
• /
• 2019

Amyloid positron emission tomography (PET) allows early and accurate diagnosis in suspected cases of Alzheimer's disease (AD) and contributes to future treatment plans. In the present study, a method of implementing a diagnostic system to distinguish ${\beta}$-Amyloid ($A{\beta}$) positive from $A{\beta}$ negative with objectiveness and accuracy was proposed using a machine learning approach, such as the Principal Component Analysis (PCA) and Support Vector Machine (SVM). $^{18}F$-Florbetaben (FBB) brain PET images were arranged in control and patients (total n = 176) with mild cognitive impairment and AD. An SVM was used to classify the slices of registered PET image using PET template, and a system was created to diagnose patients comprehensively from the output of the trained model. To compare the per-slice classification, the PCA-SVM model observing the whole brain (WB) region showed the highest performance (accuracy 92.38, specificity 92.87, sensitivity 92.87), followed by SVM with gray matter masking (GMM) (accuracy 92.22, specificity 92.13, sensitivity 92.28) for $A{\beta}$ positivity. To compare according to per-subject classification, the PCA-SVM with WB also showed the highest performance (accuracy 89.21, specificity 71.67, sensitivity 98.28), followed by PCA-SVM with GMM (accuracy 85.80, specificity 61.67, sensitivity 98.28) for $A{\beta}$ positivity. When comparing the area under curve (AUC), PCA-SVM with WB was the highest for per-slice classifiers (0.992), and the models except for SVM with WM were highest for the per-subject classifier (1.000). We can classify $^{18}F$-Florbetaben amyloid brain PET image for $A{\beta}$ positivity using PCA-SVM model, with no additional effects on GMM.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.8
• /
• pp.3865-3883
• /
• 2016

Infrastructure as a Service (IaaS) encapsulates computer hardware into a large amount of virtual and manageable instances mainly in the form of virtual machine (VM), and provides rental service for users. Currently, VM anomaly incidents occasionally occur, which leads to performance issues and even downtime. This paper aims at detecting anomalous VMs based on performance metrics data of VMs. Due to the dynamic nature and increasing scale of IaaS, detecting anomalous VMs from voluminous correlated and non-Gaussian monitored performance data is a challenging task. This paper designs an anomaly detection framework to solve this challenge. First, it collects 53 performance metrics to reflect the running state of each VM. The collected performance metrics are testified not to follow the Gaussian distribution. Then, it employs independent components analysis (ICA) instead of principal component analysis (PCA) to extract independent components from collected non-Gaussian performance metric data. For anomaly detection, it employs multi-class Bayesian classification to determine the current state of each VM. To evaluate the performance of the designed detection framework, four types of anomalies are separately or jointly injected into randomly selected VMs in a campus-wide testbed. The experimental results show that ICA-based detection mechanism outperforms PCA-based and LDA-based detection mechanisms in terms of sensitivity and specificity.

• Journal of Advanced Marine Engineering and Technology
• /
• v.25 no.5
• /
• pp.1130-1139
• /
• 2001

Springs are widely utilized in machine element. To find out stiffness of coil spring, the space beam theory using the finite element method is adopted in this paper. In three dimensional space, a space frame element is a straight bar of uniform cross section which is capable of resisting axial forces, bending moments about two principal axes in the plane of its cross section and twisting moment about its centroidal axis. The corresponding displacement degrees of freedom are twelve. The displacements of nodal points due to small increment of force are calculated by the finite element method and the calculated nodal displacements are added to coordinates of nodal points. The new stiffness matrix of the system using the new coordinates of nodal points is adopted to calculated the another increments of nodal displacements, that is, the step by step method is used in this paper. The results of the finite element method are fairly well agreed with those of various experiments. Using MATLAB program developed in this paper, spring constants can be predicted by input of few factors.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2007.11a
• /
• pp.583-588
• /
• 2007

This paper investigates the vibration characteristics of a wire-bonding transducer horn for high speed welding devices. The sample wire-bonder uses the input frequency of 136 kHz. The ultrasonic excitation causes the various vibrations of transducer horn and capillary. The vibration modes and frequencies close to the exciting frequency are identified using ANSYS. The nodal lines and amplification ratio of the ultrasonic horn are also obtained in order to evaluate the bonding performance of the sample wire-bonder system. The FEM results and experimental results show that the sample wire-bonder system uses the bending mode of 136 kHz as principal motion for bonding. The major longitudinal mode exists at 119 kHz below the excitation frequency. It is recommeded that the sample system is to set the excitation frequency at 119 kHz to improve bonding performance.

• Journal of Advanced Marine Engineering and Technology
• /
• v.32 no.2
• /
• pp.250-255
• /
• 2008

Springs are widely utilized in machine element. To find out stiffness of frustum-shaped coil spring, the space beam theory using the finite element method is adopted in this paper. In three dimensional space, a space frame element is a straight bar of uniform cross section which is capable of resisting axial forces, bending moments about two principal axes in the plane of its cross section and twisting moment about its centroidal axis. The corresponding displacement degrees of freedom are twelve. To find out load vector of coil spring subjected to distributed compression. principle of virtual work is adapted. And this theory was programming using MATLAB software. To compare FEM using MATLAB software was applied MSC. Nastran software. The geometry model for MSC. Patran was produced by 3-D design modeling software. Finite element model was produced by MSC. Patran. Finite element was applied tetra (CTETRA) having 10 node. The analysis results of the MATLAB and MSC. Nastran are fairly well agreed with those of various experiments. Using MATLAB program proposed in this paper and MSC. Nastran, spring constants and stresses can be predicted by input of few factors.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.28 no.9
• /
• pp.1352-1358
• /
• 2004

The mechanical spring is one of widely used machine elements. Among various kinds, flat-type spring loaded by a rotating pin was studied. A flat spring was simplified to a cantilever beam, and numerical analysis was attempted. Since the loading pin rotates about a separate axis from the fixed spring or vice versa, the location, direction, and magnitude of the contact force including normal contact and friction loads vary accordingly. Meanwhile, the spring is deformed substantially as the relative motion progresses. Therefore, this problem needs to be formulated taking the follower loading characteristics and geometrical non-linearity into account. Derived nonlinear differential equation was solved to yield the spring deflection, contact force and the torque to rotate the pin, and the result was compared with a finite element solution. Also, the influences of principal design parameters were studied. The proposed methodology is expected to be useful for the design of pin-loaded flat spring and the prevention of mechanical failures in the form of yielding or fatigue failure of spring or severe wear of the components.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.21 no.3
• /
• pp.92-103
• /
• 1979

The compaction of core zone of the fill dam is very important foe increasing of the Strength of soil mass and reduction of permeability of the core. The principal objects of this study are to give the construction criteria of tamping rollers and to find out the relationships between density and permeability of soil after compaction. The results in this study are summarized as follows. 1. The core zone of fill dam should be compacted more than 8 passed because the compaction effects of clayey soil increase sharply in about 8 passes of roller. 2. The coefficient of permeability (K) increases with the thickness of compaction of soil even though the density is same. 3. The effect of compaction increases with the quantity of coarse materials such as coarse sand and gravel. 4. If D values change from 100 percent to 98 percent and from 100 percent to 95 percent, K values become 2 times and 5 times of initial K value respectively. 5. The coefficient of permeability in the field soil is very high comparing with the result of laboratory test at the same 100 percent compaction ratio, but differences between both results decrease with the decrease of compaction ratio. 6. Thickness of soil layer for the compaction should be increased for heavier compaction machine. 7. In order to get the compaction ratio of 98 percent or more, 10 to 12 passes of roller is generally required with the thickness of soil from 20cm to 30cm.

• Tribology and Lubricants
• /
• v.30 no.4
• /
• pp.199-204
• /
• 2014

Various heat-treated and surface coating methods are used to mitigate abrasion in sliding machine parts. The most cost effective of these methods involves hard coatings such as diamond-like carbon (DLC). DLC has various advantages, including a high level of hardness, low coefficient of friction, and low wear rate. In practice, a supporting layer is generally inserted between the DLC layer and the steel substrate to improve the load carrying capacity. In this study, an indentation and sliding contact problem involving a small, hard, spherical particle and a DLC-coated steel surface is modeled and analyzed using a nonlinear finite element code, MARC, to investigate the influence of the supporting layer thickness on the coating characteristics and the related coating failure mechanisms. The results show that the amount of plastic deformation and the maximum principal stress decrease with an increase in the supporting layer thickness. However, the probability of the high tensile stress within the coating layer causing a crack is greatly increased. Therefore, in the case of DLC coating with a supporting layer, fatigue wear can be another important cause of coating layer failure, together with the generally well-known abrasive wear.

• Tribology and Lubricants
• /
• v.29 no.6
• /
• pp.366-371
• /
• 2013

Various heat treatment and surface coating methods have been applied to machine parts. Nowadays, diamond-like carbon (DLC) coatings are widely used because of their excellent tribological characteristics. Despite the numerous studies on DLC-coated engineering surfaces, the exact wear mechanisms related to the coating thickness and elastic modulus have not been fully examined. In this study, a sliding contact problem between a small spherical hard particle and a DLC-coated steel surface is analyzed using a nonlinear finite element code, MARC. The maximum principal stress distributions and deformed surfaces are compared for different coating thicknesses and Young's modulus values. Plastically deformed surface shapes such as a groove and torus indicate that the most dominant wear mechanism for a DLC-coated surface is abrasive wear. Fatigue wear can also play a role in a case where the coating thickness is relatively large and the elastic modulus is high.