• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.4
• /
• pp.209-216
• /
• 1996

In this paper, we propose a learning method of multi-layer perceptrons (MLPs) with 8-bit data precision. The suggested method uses the cross-entropy cost function to remove the slope term of error signal in output layer. To decrease the possibility of overflows, we use 16-bit weighted sum results into the 8-bit data with appropriate range. In the forwared propagation, the range for bit-conversion is determined using the saturation property of sigmoid function. In the backwared propagation, the range for bit-conversion is derived using the probability density function of back-propagated signal. In a simulation study to classify hadwritten digits in the CEDAR database, our method shows similar generalization performance to the error back-propagation learning with 16-bit precision.

• Environmental Sciences Bulletin of The Korean Environmental Sciences Society
• /
• v.4 no.2
• /
• pp.103-110
• /
• 2000

In order to understand the wintertime intermittent characteristics of the trans-boundary air pollutant transport observed in East Asia, a numerical simulation of the long-range transport of pollutants was applied using an atmospheric transport model(STEM-II). The numerical simulation was carried out for the entire month of January 1997 and specific atmospheric aerosol (including sulfate, nitrate, and other ion compounds0 observation data were compared from four observation sites(Cheju Island, Kanghwa Island, Dazaifu, and Fukue Island). The observation data revealed that concentration peaks were intermittently observed at 3 to 4-day intervals plus the four observation sites exhibited a very similar spatial variation. The horizontal and spatial scale of the heavily polluted air masses was analyzed based on numerical results. The mechanism of the intermittent transport of air pollutants was clearly explained by a comparison of the observed data with the numerical output. It was found that the wind pattern variations associated with the synoptic scale pressure system changes play an extremely important role in the transport of pollutants in this region.

• Journal of Sensor Science and Technology
• /
• v.17 no.2
• /
• pp.114-119
• /
• 2008

Wide dynamic range active pixel sensor(APS) using a charge pump circuit has been designed by using 2-poly 4-metal $0.35{\mu}M$ standard CMOS technology. The structure of the proposed APS is similar to the structure of the conventional 3-Tr APS. The proposed unit pixel consists of one photodiode and three MOSFETs. Using a charge pump circuit, the dynamic range of the proposed APS is increased, compared to the conventional 3-Tr APS.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.8
• /
• pp.845-850
• /
• 2009

Localization which estimates a robot's position and orientation in a given environment is very important for mobile robot navigation. Although low-cost sensors are preferred for practical service robots, they suffer from the inaccurate and insufficient range information. This paper proposes a novel approach to increasing the success rate of low-cost sensor-based localization. In this paper, both the previous and the current data obtained from the IR sensors are used for localization in order to utilize as much environment information as possible without increasing the number of sensors. The sensor model used in the monte carlo localization (MCL) is modified so that the accumulated range information may be used to increase the accuracy in estimating the current robot pose. The experimental results show that the proposed method can robustly estimate the robot's pose in indoor environments with several similar places.

• Structural Engineering and Mechanics
• /
• v.57 no.3
• /
• pp.529-541
• /
• 2016

Under the interaction between dam body, dam foundation and external environment, the dam structural behavior presents the time-varying nonlinear characteristics. According to the prototypical observations, the correct identification on above nonlinear characteristics is very important for dam safety control. It is difficult to implement the description, analysis and diagnosis for dam structural behavior by use of any linear method. Based on the rescaled range analysis approach, the algorithm is proposed to identify and extract the fractal feature on observed dam structural behavior. The displacement behavior of one actual dam is taken as an example. The fractal long-range correlation for observed displacement behavior is analyzed and revealed. The feasibility and validity of the proposed method is verified. It is indicated that the mechanism evidence can be provided for the prediction and diagnosis of dam structural behavior by using the fractal identification method. The proposed approach has a high potential for other similar applications.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.7
• /
• pp.974-988
• /
• 1995

In this paper, we propose a range image segmentation algorithm using robust regression. We derive a least $\kappa$th-order square (LKS) method by generalizing the least median of squares (LMedS) method and compare it with the conventional robust regressions. The LKS is robuster against outliers than the LMedS and shows performance similar to the residual consensus (RESC). The RESC uses the predetermined number of sorted residuals, whereas the LKS uses an adaptive parameter determined by given observations rather than the a priori knowledge. Computer simulation with synthetic and real range images shows that the proposed LKS algorithm gives better performance than the conventional ones.

• Journal of the Ergonomics Society of Korea
• /
• v.15 no.1
• /
• pp.39-52
• /
• 1996

This paper measured the range of arm joint motion for Korean 54 males in twenties. The range of the arm joint motion of the subjects was measured directly using Goniometer and protractor. The number of the static and the dynamic anthropometric variables are seven and thirteen, respectively. The anthropometric data are analyzed by basic statistical analysis (four group), correlation analysis and regression analysis using commercial SAS program. The results of analysis are compared with American students anthropometric data by Laubach(1978). Thin subjects have larger movement angle as wrist flexion, wrist abduction, elbow flexion, and elbow wupination and have smaller as wrist adduction and shoulder flexion. Fat subjects have larger movement angle as shoulder flexion and are smaller wrist abduction, elbow flexion, pronation, shoulder extension, shoulder adduction, shoulder abduction, and shoulder medial rotation Korean are more flexible than American in wrist and ranges of elbow flexion and elbow rotation. The shoulder movement is similar to that of American, but shoulder flexion is less flexible.

• Journal of Technologic Dentistry
• /
• v.34 no.3
• /
• pp.221-226
• /
• 2012

Purpose: This study was performed to compare mechanical properties for sintering methods of porous Ti implants. Methods: The specimens of Ti implant were fabricated by several sintering methods. One of them is spark plasma sintering(SPS). Another is electro discharge singering(EDS) and the other is high vacuum sintering(HVS). Mechanical properties of porous Ti implants were evaluated by universal testing machine(UTM) and their fracture surface was examined under a sanning electron microscope(SEM). Results: The tensile strength was in a range of 71 to 230 MPa, and Young's modulus was in a range of 11 to 21 Gpa. It matched with range of cortical bone. Conclusion: Mechanical properties of porous Ti implants were similar to human bone. It was shown that sintering methods of spherical powders can efficiently produce porous Ti implants with various porosities. Porous metals will be commonly used in orthopedic and dental application despite of initial focus has been on bioceramics.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.24 no.10
• /
• pp.5-12
• /
• 2010

The purpose of this study is to present the EV range in HDRI acquisition process to create luminance map. The proposed method in this study is to capture the scene at EV ${\pm}0$ that is the longest exposure points and reference point in the scene. With this reference point, sets of 25 LDRI test case were taken manually at ${\pm}2$ EV using the aperture-priority manual mode. The 25 HDRIs were created using Adobe Photoshop. The HDRIs were then imported Radiance lighting simulation program to be analyzed into falsecolor. The analysis results of the 25 HDRIs test case are 50[%] of the all tested case have a margin of error of 10[%]. In case of f/5.6, the luminance map generated with HDRI were similar to the spot luminance meter. As a result, the EV range to reduce error of luminance map generated with HDRI is EV $+2{\sim}{\pm}0{\sim}-10$.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.11 no.4
• /
• pp.116-123
• /
• 2008

In this paper we study the performance of the measurement time-delay estimation of tightly-coupled GPS/INS(Global positioning system/Inertial Navigation system) system. Generally, the heading error estimation performance of loosely-coupled GPS/INS system using GPS's Navigation Solution is poor. In the case of tightly-coupled GPS/INS system using pseudo-range and pseudo-range rate, the heading error estimation performance is better. However, the time-delay error on the measurement(pseudo-range rate) make the heading error estimation performance degraded. So that, we propose the time-delay model on the measurement and compose the time-delay estimator. And we confirm that the heading error estimation performance in the case of measurement time-delay existence is similar with the case of no-delay by Monte-Carlo simulation.