• Journal of the Korea Society of Computer and Information
• /
• v.24 no.1
• /
• pp.33-39
• /
• 2019

In this paper, we propose deep convolutional neural network(CNN) with bottleneck structure which improves the performance of earthquake classification. In order to address all possible forms of earthquakes including micro-earthquakes and artificial-earthquakes as well as large earthquakes, we need a representation and classifier that can effectively discriminate seismic waveforms in adverse conditions. In particular, to robustly classify seismic waveforms even in low snr, a deep CNN with 1x1 convolution bottleneck structure is proposed in raw seismic waveforms. The representative experimental results show that the proposed method is effective for noisy seismic waveforms and outperforms the previous state-of-the art methods on domestic earthquake database.

• Nuclear Engineering and Technology
• /
• v.55 no.2
• /
• pp.443-451
• /
• 2023

Machine Learning has introduced many solutions in data science, but its application in IoT faces significant challenges, due to the limitations in memory size and processing capability of constrained devices. In this paper we design an automatic gamma radiation detection and identification embedded system that exploits the power of TinyML in a SiPM micro radiation sensor leveraging the Edge Impulse platform. The model is trained using real gamma source data enhanced by software augmentation algorithms. Tests show high accuracy in real time processing. This design has promising applications in general-purpose radiation detection and identification, nuclear safety, medical diagnosis and it is also amenable for deployment in small satellites.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2021.11a
• /
• pp.59-60
• /
• 2021

The unit-water content has a major problem in concrete structures which leads to micro cracks on the concrete during drying time. Thus, the compressive strength and durability of the concrete structures are significantly reduced. Several techniques have been developed to measure the unit-water content in concrete structures such as heating drying, unit volume mass, and capacitance measurements. However, these techniques have problems in during measurement such as longer time, expensive and difficult in analysis of data. Frequency Domain Reflectivity (FDR) is one of the sensors which used to measure the water content. This method has several advantages including easy to measure, inexpensive, and capable of measuring moisture in real time. In this study, an attempt has been made to evaluate the unit-water content in concrete using the FDR sensor and interpret the data with deep learning method.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.17 no.4
• /
• pp.54-62
• /
• 2018

As a part of realization of artificial intelligence signal(AI Signal), this study proposed an actuated signal algorithm based on vehicle queue length that estimates in real time by deep learning. In order to implement the algorithm, we built an API(COM Interface) to control the micro traffic simulator Vissim in the tensorflow that implements the deep learning model. In Vissim, when the link travel time and the traffic volume collected by signal cycle are transferred to the tensorflow, the vehicle queue length is estimated by the deep learning model. The signal time is calculated based on the vehicle queue length, and the simulation is performed by adjusting the signaling inside Vissim. The algorithm developed in this study is analyzed that the vehicle delay is reduced by about 5% compared to the current TOD mode. It is applied to only one intersection in the network and its effect is limited. Future study is proposed to expand the space such as corridor control or network control using this algorithm.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.28 no.4
• /
• pp.55-61
• /
• 2024

Recently, the number of aging concrete structures is steadily increasing. This is because many of these structures are reaching their expected lifespan. Such structures require accurate inspections and persistent maintenance. Otherwise, their original functions and performance may degrade, potentially leading to safety accidents. Therefore, research on objective inspection technologies using deep learning and computer vision is actively being conducted. High-resolution images can accurately observe not only micro cracks but also spalling and exposed rebar, and deep learning enables automated detection. High detection performance in deep learning is only guaranteed with diverse and numerous training datasets. However, surface damage to concrete is not commonly captured in images, resulting in a lack of training data. To overcome this limitation, this study proposed a method for generating concrete surface damage images, including cracks, spalling, and exposed rebar, using stable diffusion. This method synthesizes new damage images by paired text and image data. For this purpose, a training dataset of 678 images was secured, and fine-tuning was performed through low-rank adaptation. The quality of the generated images was compared according to three base models of stable diffusion. As a result, a method to synthesize the most diverse and high-quality concrete damage images was developed. This research is expected to address the issue of data scarcity and contribute to improving the accuracy of deep learning-based damage detection algorithms in the future.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.2290-2295
• /
• 2005

In this paper, it is developed simulator system of 3 D.O.F parallel robot for educate of expertness. This simulator system is composed of three parts ? 3 D.O.F parallel robot, controller (hardware) and software. First, basic structure of the robot is 3 active rotary actuator that small geared step motor with fixed base. An input-link is connected to this actuator, and this input-link can connect two ball joints. Thus, two couplers can be connected to the input-link as a pair. An end-plate, which is jointed by a ball joint, can be connected to the opposite side of the coupler. A sub-link is produced and installed to the internal spring, and then this sub-link is connected to the upper and bottom side of the coupler in order to prevent a certain bending or deformation of the two couplers. The robot has the maximum diameter of 230 mm, 10 kg of weight (include the table), and maximum height of 300 mm. Hardware for control of the robot is composed of computer, micro controller, pulse generator, and motor driver. The PC used in the controller sends commands to the controller, and transform signals input by the user to the coordinate value of the robot by substituting it into equations of kinematics and inverse kinematics. A controller transfer the coordinate value calculated in the PC to a pulse generator by transforming it into signals. A pulse generator analyzes commands, which include the information received from the micro controller. A motor driver transfer the pulse received from the pulse generator to a step motor, and protects against the over-load of the motor Finally, software is a learning purposed control program, which presents the principle of a robot operation and actual implementation. The benefit of this program is that easy for a novice to use. Developed robot simulator system can be practically applied to understand the principle of parallel mechanism, motors, sensor, and various other parts.

• Journal of Digital Convergence
• /
• v.20 no.3
• /
• pp.415-422
• /
• 2022

This study requires a new paradigm for universities in line with the global pandemic and the 4th industrial revolution. Accordingly, we propose an educational plan for the H university online graduate school in Korea. As a research method, the implications of scholars and experts on future education were synthesized, and the cases of overseas universities using it were analyzed to propose an online graduate school education plan. As a result of the study, online graduate school needs diversity as a venue for providing new opportunities as lifelong education, and to realize this, they use a microcredit. Blockchain technology is introduced so that microcredit can be transparently verified. In addition, to correspond to various convergence major programs and further develop them, problem-solving-oriented teaching methods enhance students' convergent skills as well as active learning and interaction. More detailed curriculum research at online graduate schools is needed in the future, and we hope that you will contribute to the development of online graduate school education based on this study.

• Korea Trade Review
• /
• v.45 no.2
• /
• pp.229-262
• /
• 2020

The price-to-earnings ratio (PER) is an important indicator to measure the stock price and profitability of a firm; it is also the most used valuation indicator among investors. When using the PER to compare the investment values of different stocks, these stocks must come from the same sector. This study mainly focuses on the China's listed manufacturing firms. By learning from previous research results and analyzing the current situation, we studied the correlation between the manufacturing sector's PER and its influencing factors from both macro and micro perspectives, the combination of which eventually sheds light on such correlation. Analyzing GDP growth rate data, Manufacturing Purchasing Managers' Index, and other macroeconomic variables from 2008 to 2018, we conclude that these variables jointly have a certain impact on the average PER of the manufacturing sector. We then form panel data based on relevant (2014-2018) data gathered from 317 of China's A-listed manufacturing firms to study the impact of micro-variables on PER. By using Stata and other software to analyze the panel data, we reach the conclusion that the Debt to Asset Ratio, Return on Equity, EPS growth rate, Operating Profit Ratio, Dividend Payout Ratio, and firm size have a significant impact on PER. The Current Ratio, Treasury Stock ratio and Ownership Concentration have no distinct effect on PER. Based on our empirical findings, we design a theoretical model that affects the PER.

• Journal of Yeungnam Medical Science
• /
• v.24 no.2
• /
• pp.148-153
• /
• 2007

Background : Computer- and web-based simulation methods help students develop problem solving and decision making skills. In addition, they provide reality based learning to the student clinical experience with immediate medical feedback as well as repetitive training, on-site reviews and case closure. Materials and Methods : Seventy-five third-year medical students participated in a two-week simulation program. The students selected four modules from eight modules as follows: airway and breathing 1, cardiac arrest 1, cardiac arrhythmia 1, and chest pain 1, and then selected the first case within each of the modules. After 2 weeks, a pass score was obtained and the data analyzed. The average pass score of over 70% was considered a passing grade for each module. If the student did not pass each module, there was no score (i.e., pass score was zero). In addition, when at least one of the four modules was zero, the student was not included in this study. Results : Seventy-five students participated in the simulation program. Nineteen students were excluded based on their performance. The final number of students studied was 56 students (74.7%). The average scores for each module 1 to 4 were 86.7%, 85.3%, 84.0%, and 84.0%, and the average obtained pass score was 88.6 for the four modules in all 56 students. Conclusion : Medical simulation enabled students to experience realistic patient situations as part of medical learning. However, it has not been incorporated into traditional educational methodology. Here we describe the introduction and the development of various simulation modules and technologies for medical education.

• Journal of The Korean Association of Information Education
• /
• v.15 no.4
• /
• pp.571-578
• /
• 2011

In this study, we suggested a strategy about a discussion class using Twitter, and experimented it inside an elementary school classroom. Elementary students participated in a panel discussion and the others discussed as audience using Twitter. After the discussion, we investigated the effectiveness of our strategy using the Technology Acceptance Model and verified students' satisfaction and ability to collaborate through giving them a questionnaire. As a result, the perceived ease of use positively effected the perceived usefulness and the perceived usefulness influenced the attitude and the attitude affect on intention to use. Also, students were satisfied with the discussion class on Twitter and had a positive perception about collaboration with it. As a result of regression, perception of collaboration among the students influenced the perceived usefulness positively. The results in this study show the effectiveness of using the discussion class strategy on Twitter.