• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.111-114
• /
• 2000

척수 손상에 의한 사지마비 중증 장애인이 일상 생활을 함에 있어서 가족이나 주변인의 도움을 받지 않고서는 가전제품의 작동이나 전화 통화 등과 같은 작은 일 조차 스스로 할 수 없는 현실에서 국내 여건에 적합한 생활환경 제어장치의 개발이 필요하나 현재 이를 위한 마땅한 제어장치가 없는 실정이다. 각 가정에 널리 보급되어 있는 PC를 활용할 수 있도록 PC기반의 음성인식기를 개발하고 이를 응용한 가전제품 제어기를 개발, 제작하여 사지마비 환자 스스로 간단한 동작만으로 환자와 가족의 신체적, 정신적 부담을 감소 하고자 한다 이장치는 음성인식기 부분과 제어장치 부분 그리고 I/O 인터페이스 부분으로 구성 되어있다. 중증장애인은 이 장치를 사용하여 일상생활에서 기본적인 자유로운 생활의 영위를 도모하며 환자 가족의 부담을 줄일 수 있으며 환자와 가족 모두 심리적인 만족을 얻은 것으로 나타난다. 그러나 기기 사용을 위한 환자의 음성인식을 위한 언어학습과정의 불편함과 인식에러의 과제가 남아있다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.319-319
• /
• 2020

최근 데이터 과학의 획기적인 발전 덕분에 딥러닝 (Deep Learning) 알고리즘이 개발되어 다양한 분야에 널리 적용되고 있다. 본 연구에서는 인공신경망 중 하나인 LSTM(Long-Short Term Memory) 네트워크를 사용하여 댐 유입량을 예측하였다. 구체적인 내용으로, (1) LSTM에 필요한 입력 데이터를 효율적으로 사전 처리하는 방법, (2) LSTM의 하이퍼 매개변수를 결정하는 방법 및 (3) 다양한 손실 함수(Loss function)를 선택하고 그 영향을 평가하는 방법 등을 다루었다. 제안된 LSTM 모델은 강우량(R), 댐유입량(Q) 기온(T), 기저유량(BF) 등을 포함한 다양한 입력 변수들의 함수로 가정하였으며, CCF(Cross Correlations), ACF(Autocorrelations) 및 PACF(Partial Autocorrelations) 등의 기법을 사용하여 입력 변수를 결정하였다. 다양한 sequence length를 갖는 (즉 t, t-1, … t-n의 시간 지연을 갖는) 입력 변수를 적용하여 데이터 학습에 최적의 시퀀스 길이를 결정하였다. LSTM 네트워크 모델을 적용하여 2014년부터 2020년까지 한강 유역 9개의 댐 유입량을 추정하였다. 본 연구로부터 댐 유입량을 예측하는 것은 홍수 및 가뭄 통제를 위한 필수 요건들 중 하나이며 수자원 계획 및 관리에 도움이 될 것이다.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.1
• /
• pp.40.2-40.2
• /
• 2021

최근 COVID-19의 확산에 따라 비대면 학습의 중요성이 화두로 오르고 있다. 과학관과 같은 다중이용시설을 이용하는 방법에서 각 학교, 또는 가정에서 지도할 수 있는 교육 방법의 변화 또한 논의해야 할 때이다. 특히 천문학 교육의 장으로 활용되고 있는 천체투영관(Planetarium)은 다중이용시설로, 상황에 따라 여러 가지 제약을 가지므로 시·공간 제약을 받지 않는 교육프로그램의 개발이 필요하고 볼 수 있다. 본 연구에서는 천체시뮬레이션 소프트웨어를 활용한 스크립트 제작 및 360° VR 영상 제작기술을 개발하고 온라인 플랫폼을 활용하여 천체투영관을 방문하지 않은 비대면 환경에서 도 참가자개인의 휴대기기등을 통한 몰입도 높은 천문학 교육을 구현할 수 있는 프로그램을 개발하여 새로운 환경에 대한 적용 기회를 제공하고자 한다. 나아가 개발한 프로그램을 천체투영관에 적용하여 천체투영관 활용 천문학 교육프로그램과 교육과정과 연계한 교육콘텐츠 개발에 활용되기를 기대한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.6
• /
• pp.720-725
• /
• 2004

Robot systems have applied to manufacturing or industrial field for reducing the need for human presence in dangerous and/or repetitive tasks. However, robot applications are transformed from industrial field to human life in recent tendency Nowadays, final goal of robot is to make a intelligent robot that can understand what human say and learn by itself and have internal emotion. For example Home service robots are able to provice functions such as security, housework, entertainment, education and secretary To provide various functions, home robots need to recognize human`s requirement and environment, and it is indispensable to use artificial intelligence technology for implementation of home robots. In this paper, implemented robot system takes data from several sensors and fuses the data to recognize environment information. Also, it can select a proper behavior for environment using soft computing method. Each behavior is composed with intuitive motion and sound in order to let human realize robot behavior well.

• Smart Media Journal
• /
• v.11 no.9
• /
• pp.9-20
• /
• 2022

Deep reinforcement learning(RL) is an end-to-end data-driven control method that is widely used in the autonomous driving domain. However, conventional RL approaches have difficulties in applying it to autonomous driving tasks due to problems such as inefficiency, instability, and uncertainty. These issues play an important role in the autonomous driving domain. Although recent studies have attempted to solve these problems, they are computationally expensive and rely on special assumptions. In this paper, we propose a new algorithm MCDT that considers inefficiency, instability, and uncertainty by introducing a method called uncertainty sequence modeling to autonomous driving domain. The sequence modeling method, which views reinforcement learning as a decision making generation problem to obtain high rewards, avoids the disadvantages of exiting studies and guarantees efficiency, stability and also considers safety by integrating uncertainty estimation techniques. The proposed method was tested in the OpenAI Gym CarRacing environment, and the experimental results show that the MCDT algorithm provides efficient, stable and safe performance compared to the existing reinforcement learning method.

• Journal of Korean Home Economics Education Association
• /
• v.31 no.4
• /
• pp.81-95
• /
• 2019

This study examined the research trends of sustainable development education in the family life area, focusing on papers published in journals specialized in the Practical Arts (Technology & Home Economics) subject from 1989 to 2018. The papers from three journals were selected based on the titles, abstracts, and keywords for their contents related to sustainable development education. The results were analyzed by frequency analysis and cross-tabulation. Sustainable development education has surged since the mid-2000s when national interest on this topic sharply increased. Rather than focusing on particular content areas in home economics, most research was conducted through an integrated approach. More than half of the studies were conducted in experimental studies, with the largest number of studies involving elementary school students. Of the three components of ESD, research on society was most frequently conducted, followed by the environmental area. The contents of the home economics curriculum and the details of sustainable development were related. More related research should be conducted so that various sustainable development education can improve students' practical problem-solving ability through Practical Arts (Technology & Home Economics) curriculum.

• Journal of Korean Home Economics Education Association
• /
• v.20 no.3
• /
• pp.145-161
• /
• 2008

The purpose of this paper is to find out the needs of housing contents in highschool Home Economics among students. The 105 housing contents and topics were chosen from the 3 textbooks and related references. The number of 516 questionnaire from 8 areas were used to analyze. Analyses indicated that respondents were more likely to favor Home Economics subject and one out of five wanted to select it. Average needs of the housing contents was above middle(3.4) and these was a little difference among 3 middle units in housing. The most needed unit was housing design. There was a more wide difference in 19 small units than in middle ones, ranged $3.00\sim3.91$. Individual contents showed the most variousness in needs from 2.89 to 4.30 in the middle unit of housing and environments, 3.00 to 4.01 in housing design, and 2.68 to 3.91 in housing management. The needs of additional contents were about the same as those textbook contents and it indicated that new textbook needs to adopt the newly demanded housing contents. Some of individual, family, and environmental characteristics were significantly related to the students needs. The higher positive attitude such as favor, interest, and selecting of the subject was related to the higher needs of housing contents. Residential areas and neighboring were also related to most of the needs of individual housing contents.

• The Journal of Korean Association of Computer Education
• /
• v.13 no.2
• /
• pp.25-34
• /
• 2010

The Internet has brought us a new paradigm called WEB 2.0 which inherently represents the openness, participation, sharing, and cooperation. The WEB 2.0 has rendered people to actively participate in the creation of information and to dynamically interact with others to build mutual knowledge-bases, introducing a whole new web environment. One of the most representative techniques demonstrating the value of the WEB 2.0 is the WIKI which is essentially based on the 'Collective Intelligence' and the 'Wisdom of Crowds'. So far, the WIKI has drawn lots of attention for its potential as an educational tool. In this research, we explored the educational potential of the WIKI by investigating various programs and web-based tools offering WIKI services, and then, analyzing the usage model and characteristics of its users. Based on the analysis, we developed a WIKI-based instructional model and also proposed concrete lesson plans adopting this model. Besides, we introduced preliminary methodologies on the active use of the WIKI in the Edunet and the Cyber Home Learning System as well.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.6
• /
• pp.110-121
• /
• 2016

The primary aim of the color constancy algorithm is to estimate illuminant chromaticity. There are various statistical-based, learning-based and combinational-based color constancy algorithms already exist. However, the statistical-based algorithms can only perform well on images that satisfy certain assumptions, learning-based methods are complex methods that require proper preprocessing and training data, and combinational-based methods depend on either pre-determined or dynamically varying weights, which are difficult to determine and prone to error. Therefore, this paper presents a new optimization based illuminant estimation method which is free from complex preprocessing and can estimate the illuminant under different environmental conditions. A strong color cast always has an odd standard deviation value in one of the RGB channels. Based on this observation, a cost function called the degree of illuminant tinge(DIT) is proposed to determine the quality of illuminant color-calibrated images. This DIT is formulated in such a way that the image scene under standard illuminant (d65) has lower DIT value compared to the same scene under different illuminant. Here, a swarm intelligence based particle swarm optimizer(PSO) is used to find the optimum illuminant of the given image that minimizes the degree of illuminant tinge. The proposed method is evaluated using real-world datasets and the experimental results validate the effectiveness of the proposed method.

• Science of Emotion and Sensibility
• /
• v.14 no.3
• /
• pp.395-402
• /
• 2011

As the consumer market for odor products grows, companies producing healthcare products are beginning to pay more attention to the emotional aspect of an odor product in order to differentiate their products from competitors. In the following research, the affective effect of odor product was investigated while focusing on relaxation and working contexts using orange and pine scents, since these are typical odors in current domestic market. Two empirical studies were carried out. First, in experiment I, 18 subjects, all of whom were university students, spent 20 minutes sitting comfortably on a sofa while electrocardiogram assessments were made. After a five-minute break, in experiment II, the same subjects were provided with both arithmetic and geometric questions and their electroencephalogram readings was recorded from eight channels. All subjects participated in three sessions - no odor, an orange scent, and then a pine scent - with a minimum time interval of 24 hours. The results show that in the context of a pine scent, both the activation ratio of subjects' parasympathetic system and those of the Sensory Motor Rhythm waves and Mid Beta waves were at the highest peak. Therefore, the pine scent helped the subjects to feel more comfortable and more focused at the same time. In other words, it gave them a state of meditated attention. In addition, it was found that the right brain was activated twice the intensity when the subjects worked through the geometric questions, whereas both sides of the brain were activated in equal magnitude during the process of arithmetic tasks. This replicates previous studies of the functional aspect of the right brain - being responsible for spatial and creative thinking.