• Journal of Digital Convergence
• /
• v.15 no.8
• /
• pp.391-400
• /
• 2017

The purpose of this paper is to present the effect of image stimulation on body stability as a conceptual model and to investigate the effect of image stimulus(2D, VR) on body stability(hand stability) through experiments Recently, stereoscopic images such as virtual and augmented reality are combined with smart phones and exercise equipments, and the diffusion is becoming active. The possibility of a safety accident or human error is also increasing as it temporarily affects the balance of the body and hand stability after the image stimulus is removed. The conceptual model is presented based on the results of previous studies. Based on the experimental results, the conceptual model has been explained in combination with the human information processing process and cognitive resource models that take place in the brain. Twenty subjects were exposed to 2D and VR stimuli, and display fatigue was measured by cybersickness questionnaire and hand stability by hand steadiness tester. Experimental results show that VR images induce higher display fatigue and lower hand stability than 2D. In this study, it is meaningful that hand stability according to image type and display fatigue level which have not been tried yet is revealed through conceptual model and experiment.

• Journal of Digital Convergence
• /
• v.18 no.3
• /
• pp.163-169
• /
• 2020

A smart key has been used in a variety of embedded environments and there also have been attacks from a remote place by amplifying signals at a location of a user. Existing studies on defence techniques suggest multiple sensors and hash functions to improve authentication speed; these, however, increase the electricity usage and the probability of type 1 error. For these reasons, I suggest an embedded security controller based on client authentication and user movement information improving the authentication method between a controller and a host device. I applied encryption algorithm to the suggested model for communication using an Arduino board, GPS, and Bluetooth and performed authentication through path analysis utilizing user movement information for the authentication. I found that the change in usability was nonsignificant when performing actions using the suggested model by evaluating the time to encode and decode. The embedded security controller in the model can be applied to the system of a remote controller for a two-wheeled vehicle or a mobile and stationary host device; in the process of studying, I found that encryption and decryption could take less then 100ms. The later study may deal with protocols to speed up the data communication including encryption and decryption and the path data management.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.4
• /
• pp.313-323
• /
• 2021

The purpose of this study is to propose rules that can identify the speaker referred by the speaker directive in the text for the realization of a machine learning-based virtual character using a literary text. Through previous studies, we found that when applying literary texts to machine learning, the machine did not properly discriminate the speaker without any specific rules for the analysis of speaker directives such as other names, nicknames, pronouns, and so on. As a way to solve this problem, this study proposes 'nine rules for finding a speaker indicated by speaker directives (including pronouns)': location, distance, pronouns, preparatory subject/preparatory object, quotations, number of speakers, non-characters directives, word compound form, dispersion of speaker names. In order to utilize characters within a literary text as virtual ones, the learning text must be presented in a machine-comprehensible way. We expect that the rules suggested in this study will reduce trial and error that may occur when using literary texts for machine learning, and enable smooth learning to produce qualitatively excellent learning results.

• The Korean Journal of Applied Statistics
• /
• v.26 no.5
• /
• pp.847-856
• /
• 2013

We propose a modified learning process for generalized neural network using a learning algorithm by Liu et al. (2001). We consider the effect of initial weights, training results and learning errors using a modified learning process. We employ an incremental training procedure where training patterns are learned systematically. Our algorithm starts with a single training pattern and a single hidden layer neuron. During the course of neural network training, we try to escape from the local minimum by using a weight scaling technique. We allow the network to grow by adding a hidden layer neuron only after several consecutive failed attempts to escape from a local minimum. Our optimization procedure tends to make the network reach the error tolerance with no or little training after the addition of a hidden layer neuron. Simulation results with suitable initial weights indicate that the present constructive algorithm can obtain neural networks very close to minimal structures and that convergence to a solution in neural network training can be guaranteed. We tested these algorithms extensively with small training sets.

• Journal of Digital Convergence
• /
• v.10 no.2
• /
• pp.243-247
• /
• 2012

Beam hardening artifact happens in the CT image. when a PET/CT is conducted while there is a metallic dental implant. The artifact appears in the CT image can affect the PET image. When the patient with head and neck cancer has a metallic dental implant, Beam hardening artifact which was taken in th CT image can change the PET image and SUV value. Therefore, by Quantitative measure of the SUV according to the change in HU by the metallic dental implant, the appropriacy in the clinical application was assessed. The records of 47 patients with PET/CT August 2011. For the analysis, 2 region of interest were defined in area where CT and PET image. As a result of the experiment, if there in an implant, the HU and the SUV increased and there existed a statistically significant difference(p<0.01). Although this level of increase was not large compared with that in the patient who have no metallic dental implant, when a person has head and neck cancer, it is even more likely to be overestimated when diagnosing the cancer. When conducting PET/CT for the patient who have head and neck cancer, the physical biological parts should be considered in order not to make an error in decoding.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.2
• /
• pp.19-24
• /
• 2020

The interest in renewable energy is increasing due to the depletion of fossil fuels. In particular, active research on wave power, which is highly predictable and abundant, is being conducted. The coaxial accelerator-type wave power generator used in this study was designed to improve the power generation efficiency by converting bidirectional linear motion into a rotational force. In an offshore engineering basin, waves were generated, and case tests were performed according to the wave period and wave height. The experimental results were verified by the theoretical method related to the frequency response, and the overall trend was confirmed to be consistent. These results are expected to be useful in estimating the power of wave generators and designing parameters to improve the efficiency of wave energy in the design stage before manufacturing. In addition, the manufacturer can predict the wave energy efficiency of wave generators, which can reduce the development time and cost by preventing trial and error processes.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.6
• /
• pp.137-143
• /
• 2015

A delay-line type time-to-digital converter (TDC) implemented in a field-programmable gate array (FPGA) is most widely owing due to its simple structure and high conversion rate. However, the delay-line type TDC suffers from nonlinearity error caused by the long delay-line because its time interval measurement range is determined by the length of the used delay line. In this study, a new TDC structure with a shorter delay line by taking a pulse train as an input is proposed for improved time accuracy and efficient use of resources. The proposed TDC utilizes a pulse-train with four transitions and a transition state detector that identifies the used transition among four transitions and prevents the meta-stable state without a synchronizer. With 72 delay cells, the measured resolution and maximum non-linearity were 20.53 ps, and 1.46 LSB, respectively, and the time interval measurement range was 5070 ps which was enhanced by approximately 343 % compared to the conventional delay-line type TDC.

• Journal of Korea Water Resources Association
• /
• v.51 no.11
• /
• pp.1021-1029
• /
• 2018

A primary objective of this study is to develop a drought forecasting technique based on groundwater which can be exploit for water supply under drought stress. For this purpose, we explored the lagged relationships between regionalized SGI (standardized groundwater level index) and SPI (standardized precipitation index) in view of the drought propagation. A regional prediction model was constructed using a NARX (nonlinear autoregressive exogenous) artificial neural network model which can effectively capture nonlinear relationships with the lagged independent variable. During the training phase, model performance in terms of correlation coefficient was found to be satisfactory with the correlation coefficient over 0.7. Moreover, the model performance was described by root mean squared error (RMSE). It can be concluded that the proposed approach is able to provide a reliable SGI forecasts along with rainfall forecasts provided by the Korea Meteorological Administration.

• Korean Journal of Remote Sensing
• /
• v.37 no.2
• /
• pp.221-232
• /
• 2021

This study analyzed co-registration results according to the geometric processing level of reference image, which are Level 1R and Level 1G provided from KOMPSAT-3 and KOMPSAT-3A images. We performed co-registration using each Level 1R and Level 1G image as a reference image, and Level 1R image as a sensed image. For constructing the experimental dataset, seven Level 1R and 1G images of KOMPSAT-3 and KOMPSAT-3A acquired from Daejeon, South Korea, were used. To coarsely align the geometric position of the two images, SURF (Speeded-Up Robust Feature) and PC (Phase Correlation) methods were combined and then repeatedly applied to the overlapping region of the images. Then, we extracted tie-points using the SURF method from coarsely aligned images and performed fine co-registration through affine transformation and piecewise Linear transformation, respectively, constructed with the tie-points. As a result of the experiment, when Level 1G image was used as a reference image, a relatively large number of tie-points were extracted than Level 1R image. Also, in the case where the reference image is Level 1G image, the root mean square error of co-registration was 5 pixels less than the case of Level 1R image on average. We have shown from the experimental results that the co-registration performance can be affected by the geometric processing level related to the initial geometric relationship between the two images. Moreover, we confirmed that the better geometric quality of the reference image achieved the more stable co-registration performance.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.30 no.3A
• /
• pp.193-200
• /
• 2010

Structural design requires simultaneously to ensure safety by considering quantitatively uncertainties in the applied loadings, material properties and fabrication error and to maximize economical efficiency. As a solution, system reliability-based design optimization (SRBDO), which takes into consideration both uncertainties and economical efficiency, has been extensively researched and numerous attempts have been done to apply it to structural design. Contrary to conventional deterministic optimization, SRBDO involves the evaluation of component and system probabilistic constraints. However, because of the complicated algorithm for calculating component reliability indices and system reliability, excessive computational time is required when the large-scale finite element analysis is involved in evaluating the probabilistic constraints. Accordingly, an algorithm for SRBDO exhibiting improved stability and efficiency needs to be developed for the large-scale problems. In this study, a more stable and efficient SRBDO based on the performance measure approach (PMA) is developed. PMA shows good performance when it is applied to reliability-based design optimization (RBDO) which has only component probabilistic constraints. However, PMA could not be applied to SRBDO because PMA only calculates the probabilistic performance measure for limit state functions and does not evaluate the reliability indices. In order to overcome these difficulties, the decoupled algorithm is proposed where RBDO based on PMA is sequentially performed with updated target component reliability indices until the calculated system reliability index approaches the target system reliability index. Through a mathematical problem and ten-bar truss problem, the proposed method shows better convergence and efficiency than other approaches.