• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.37 no.2
• /
• pp.39-49
• /
• 2000

In this paper, we propose an optimizing feature extraction method for multiclass problems assuming normal distributions. Initially, We start with an arbitrary feature vector Assuming that the feature vector is used for classification, we compute the classification error Then we move the feature vector slightly in the direction so that classification error decreases most rapidly This can be done by taking gradient We propose two search methods, sequential search and global search In the sequential search, an additional feature vector is selected so that it provides the best accuracy along with the already chosen feature vectors In the global search, we are not constrained to use the chosen feature vectors Experimental results show that the proposed algorithm provides a favorable performance.

• Journal of Ocean Engineering and Technology
• /
• v.15 no.3
• /
• pp.100-106
• /
• 2001

In this study, CT specimens were prepared from Pressure Vessel Steel which was used for pressure vessel plates for room and low temperature service. And we got the following characteristics from fatigue crack growth test carried out in the environment of room and low temperature at $25^{\circ}C$, -3$0^{\circ}C$, -6$0^{\circ}C$, -10$0^{\circ}C$ and -12$0^{\circ}C$ and in the range of stress ratio of 0.05 and 0.3 by means of opening mode displacement. At the constant stress ratio, the threshold stress intensity factor range ${\Delta}K_{th}$ in the early stage of fatigue crack growth (Region I) and stress intensity factor range ${\Delta}K$ in the stable of fatigue crack growth (Region II) was increased in proportion to descent temperature. It assumed that the fatigue resistance characteristics and fracture strength at low temperature is considerable higher than that of room temperature in the early stage and stable of fatigue crack growth region. The straight line slope relation of logarithm da/dN $-{\Delta}K$ in Region II, that is, the fatigue crack growth exponent m increased with descending temperature at the constant stress ratio. It was assumed that the fatigue crack growth rate da/dN in proportion to descending temperature in Region II and the cryogenic-brittleness greatly affect a material with decreasing temperature.

• Journal of the Korean Data and Information Science Society
• /
• v.8 no.2
• /
• pp.163-171
• /
• 1997

Function approximation from a set of input-output pairs has numerous applications in scientific and engineering areas. Multilayer feedforward neural networks have been proposed as a good approximator of nonlinear function. The back propagation(BP) algorithm allows multilayer feedforward neural networks to learn input-output mappings from training samples. It iteratively adjusts the network parameters(weights) to minimize the sum of squared approximation errors using a gradient descent technique. However, the mapping acquired through the BP algorithm may be corrupt when errorneous training data we employed. When errorneous traning data are employed, the learned mapping can oscillate badly between data points. In this paper we propose a robust BP learning algorithm that is resistant to the errorneous data and is capable of rejecting gross errors during the approximation process, that is stable under small noise perturbation and robust against gross errors.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.17 no.2 s.105
• /
• pp.158-164
• /
• 2006

In this paper, we proposed a new algorithm of the inverse scattering for the reconstruction of unknown dielectric scatterers using the finite-difference time-domain method and the design sensitivity analysis. We introduced the design sensitivity analysis based on the gradient information for the fast convergence of the reconstruction. By introducing the adjoint variable method for the efficient calculation, we derived the adjoint variable equation. As an optimal algorithm, we used the steepest descent method and reconstructed the dielectric targets using the iterative estimation. To verify our algorithm, we will show the numerical examples for the two-dimensional $TM^2$ cases.

• Environmental Mutagens and Carcinogens
• /
• v.21 no.2
• /
• pp.77-81
• /
• 2001

Kamijadowhan (KMD), an oriental herbal medicine used for anti-angiogenic effect, was extracted with 80% ethanol from mixture of source materials and lyophilized. KMD was orally administered to plugpositive pregnant rats from gestational days 12 to 20, dividing into three groups including vehicle-treated control, 0.5 g/kg or 3 g/kg KMD-treated groups. Dam weight during gestation and post-gestation, weight of pre- and post-weaning offsprings in male and female, and reproductive and developmental endpoints including incisor eruption, eye opening and testes descent were measured. No significant alterations in development of physical landmarks in offspring, maternal weight gain during gestation and post-gestation, and offspring weight were observed in KMD-treated group. The measurement of organ weight at post-gestational days 21 was not changed in dams. In 0.5 g/kg KMD-treated rats, kidney weights in male and female offsprings were significantly increased, and the body weight in male offspring was also increased. Liver and brain weights were not changed. Taken together, these data suggest that KMD may not significantly cross the placenta and produce no reproductive and developmental toxicity at maternally non-toxic dosages.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.44 no.4 s.316
• /
• pp.86-92
• /
• 2007

A discrete dynamic model for tracking feature points in 2D images is developed. Conventional snake approaches deform a contour to lock onto features of interest within an image by finding a minimum of its energy functional, composed of internal and external forces. The neighborhood around center snaxel is a space matrix, typically rectangular. The structure of the model proposed in this paper is a set of connected vertices. Energy model is designed for its local minima to comprise the set of alternative solutions available to active process. Block on tracking is one dimension, line type. Initial starting points are defined to the satisfaction of indent states, which is then automatically modified by an energy minimizing process. The track is influenced by curvature constraints, ascent/descent or upper/lower points. The advantages and effectiveness of this layer approach may also be applied to feature points tracking of digital image whose pixels have one directional properties with high autocorrelation between adjacent data lines, vertically or horizontally. The test image is the ultrasonic carotid artery image of human body, and we have verified its effect on intima/adventitia starting points tracking.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.41 no.11
• /
• pp.855-864
• /
• 2013

Active tab is one of the promising technology for the BVI (blade-vortex interaction) noise reduction, and analysis of noise reduction performance is very important phase of the technology development. For the purpose of analysing the performance of noise reduction using active tab, CAMRAD II model for a model-scale rotor system was constructed utilizing structural design result and airfoil aerodynamic data generated by CFD (computational fluid dynamics) calculation. HHC strategy was applied to descent flight condition and air-load was calculated by CAMRAD II then variations of BVI noise was calculated by in-house program. Calculation result with respect to tab length and control phase changes showed BVI noise could be reduced by -3.3dB.

• Fire Science and Engineering
• /
• v.29 no.5
• /
• pp.51-56
• /
• 2015

In this study, optimal design methods were applied to the agent discharge flow of clean agent fire extinguishing systems. The methods combined optimal design theory and engineering theory for engineering analysis in a design program or coast savings in value engineering. Optimal design parameters were determined to optimize the agent discharge flow based on the design theory of the clean agent fire extinguishing systems and the theory of optimal design. The design factors were verified in regard to suitability for the performance of fire extinguishing systems. The results provide a foundation for optimal design method methods in other fire extinguishing systems. Optimization of the agent discharge flow of the discharge nozzle was confirmed by the constraints on the inner diameter of the discharge nozzle and the pipe, agent arrival time, flow, and pressure variation of the agent. The deviation of discharge pressure and agent time of the agent discharge nozzle were found to correlate with the pressure variation.

• Journal of Broadcast Engineering
• /
• v.25 no.2
• /
• pp.157-165
• /
• 2020

In this paper. we propose an effective compensation scheme to the quantization error arisen from quantized learning in a machine learning on an embedded system. In the machine learning based on a gradient descent or nonlinear signal processing, the quantization error generates early vanishing of a gradient and occurs the degradation of learning performance. To compensate such quantization error, we derive an orthogonal compensation vector with respect to a maximum component of the gradient vector. Moreover, instead of the conventional constant learning rate, we propose the adaptive learning rate algorithm without any inner loop to select the step size, based on a nonlinear optimization technique. The simulation results show that the optimization solver based on the proposed quantized method represents sufficient learning performance.

• Journal of Advanced Navigation Technology
• /
• v.20 no.4
• /
• pp.275-284
• /
• 2016

The potential impact of aircraft emissions on the current and projected climate of our planet is one of the more important environmental issues facing the aviation industry. Increasing concern over the potential negative effects of greenhouse gas emissions has motivated aircraft emission estimation and prediction as one of the ways to reduce aircraft emissions and mitigate the impact of aviation on climate. We obtained airline flight schedules for all the airports in Korea that are included in OAG data. Fuel burn and emission index of LTO flight which contains take off, climb and approach under 3000ft and Non LTO flight which contains climb, cruise and descent over 3000ft for all the airports in Korea in 2005 were estimated and analysed for each condition using AEIC software which has been developed by MIT Lab for Aviation and Environment.