• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.06a
• /
• pp.107-108
• /
• 2017

횡단보도 인근에서는 보행취약자의 사고가 끊이지 않고 있으며 사고예방 및 사고의 절감을 위하여 선제적안 안전시스템의 개발이 요구되고 있다. 선제적 안전시스템의 개발을 위하여 빅데이터를 이용한 안전 데이터 도출, 영상분석을 이용한 보행자 행동특성 모니터링 시스템의 개발 및 사고감소를 위한 안전 시스템 개발이 진행되고 있다. 보행취약자 위험상황 판단에 대한 정의를 빅데이터 분석을 통해 도출하고 횡단보도 주변 안전 시스템의 개발을 기존 시스템에 적용 및 새로운 시스템을 개발하며 이에 적합한 딥러닝 영상분석 시스템을 개발하였다. 본 논문에서는 딥러닝 모델을 이용하여 객체의 검출, 분석을 수행하는 객체 검출부, 객체의 포즈와 행동을 보여주는 영상 분석부로 구성되어 있으며 기존 모델을 응용하여 최적화한 모델을 적용하였다. 딥러닝 모델의 구동은 리눅스 서버에서 운용되고 있으며 딥러닝 모델 구동을 위한 여러 툴을 적용하였다. 본 연구를 통하여 보행취약자의 검출, 추적, 보행취약자의 포즈 및 위험상황을 인식하고 안전시스템과 연계할 수 있도록 구성하였다.

• 대한공업교육학회지
• /
• v.37 no.1
• /
• pp.183-204
• /
• 2012

This study was to analysis perception about Teaching & Learning Support System of the school of Rural Areas in Korea. In order to accomplish the study purpose, the study conducted a research through Web Survey method and examined their recognition. Major findings of this study were as follows: First, Student was satisfied with TLSS of the school of Rural Areas in Korea. Students would like to 'using e-learning with Electronic board and Tablet PC' among program contents. But the system' utilization was low. However, 90% of the students wanted that teacher usually using the FASTEL. Second, Teacher satisfaction was lower compared to the students is relatively. Teachers prefer 'various after school program' to 'using e-learning with Electronic board and Tablet PC'. However, learning effective of using FASTEL system was a 73.8% of teachers are effective is aware and had. Third, 35% Students recognized that are using the pastel system, frequently, On the other hand, the teachers are 60% and to recognize that there was. The reason doesn't the system to good use is slow wireless. Still, e-learning classroom infrastructure for satisfaction was high. Fourth, Teachers recognized students and the pastel system interaction than other e-learning system high. FASTEL System used in the most uncomfortable things on e-learning classroom Wireless Internet phenomena and lost content lacking, in order, I was told. Therefore, there is an urgent need to improve to include Wireless Internet speed improvements and various content added support, easy-to-use instructional tools Add support order selected.

• The Journal of the Korea Contents Association
• /
• v.10 no.7
• /
• pp.449-458
• /
• 2010

The e-Learning quality assurance is a part of responsibility for education in latest information age, and should be understood as a basic direction for qualitative enhancement rather than quantifiable expansion. In such a context, this study sought to survey latest trends of e-Learning quality assurance activities deployed around overseas countries, so that it could give possible implications on how to plan and perform e-Learning quality assurance in Korea. So, this study focused on surveying actual conditions, coverage, target and useful indices of e-Learning quality assurance activities deployed by 9 institutions across 5 countries, and characterizing those activities across 5 countries. As a result, this study could find out its implications such as acquisition of various resources available for e-Learning, nationwide consolidation of e-Learning quality assurance activities, development of specialists in e-Learning quality assurance, and choice and focus for e-Learning quality assurance system. In other words, it was found that e-Learning quality assurance should be approached from nationwide standpoint in the interest of better educational quality, rather than from viewpoint of task.

• Journal of the Korea Convergence Society
• /
• v.10 no.10
• /
• pp.283-289
• /
• 2019

This study classifies the characteristics of e-learning learners into personal and psychological factors and examines how they affect the e-learning effects and intention to reuse. For this purpose, a research study was conducted targeting university students at Y and K universities in Gyeongsangbuk-do. As a result, first, the grade, self-control, concentration and self-efficacy of the learner influenced the learning effect of e-learning. Second, learner's grade, e-learning experience, self-efficacy, and learning effect were found to influence e-learning intention. Therefore, the system design and the managerial role of the instructor should be considered in order to develop various contents considering the characteristics of university students and to manage the learning process by themselves. The results of this study are expected to provide effective data for improving the learning effect of e-learning and to provide basic data on individual characteristics of e-learning learners.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.9
• /
• pp.370-378
• /
• 2012

This paper proposes a method of constructing of hybrid web-based smart learning system to operable in a variety of mobile devices. To do this, the proposed system is developed a learning system with standardized and enhanced functions. In the proposed method, API specifications based on the standard functionality of smart learning system are created. And then, by building the API provider on a legacy system an organic linkage between the legacy system and the smart learning system is guaranteed. A standard API method is applied to data integration between the PC-based learning system and the smart learning system. The smart learning system interacts with legacy learning systems though Json/XML data forms via the https protocol. As a result, the legacy system using the proposed method dose not require major modifications and changes for a smart learning service.

• Review of KIISC
• /
• v.29 no.3
• /
• pp.44-50
• /
• 2019

최근 딥러닝 기술은 다양한 분야에서 놀라운 성능을 보여주고 있어 많은 서비스에 적용되고 있다. 얼굴인식 또한 딥러닝 기술을 접목하여 높은 수준으로 얼굴인식이 가능해졌다. 하지만 딥러닝 기술은 원본 이미지를 최소한으로 변조시켜 딥러닝 모델의 오인식을 발생시키는 적대적 예제에 취약하다. 이에 따라, 본 논문에서는 딥러닝 기반 얼굴인식 시스템에 대해 적대적 예제를 이용하여 기만공격 실험을 수행하였으며 실제 얼굴에 분장할 수 있는 영역을 고려하여 설정된 변조 영역에 따른 기만공격 성능을 분석한다.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.1
• /
• pp.92-98
• /
• 2019

In this paper, we propose a deep learning system based on morphological neural network(MNN). The deep learning layers are morphological operation layer, pooling layer, ReLU layer, and the fully connected layer. The operations used in morphological layer are erosion, dilation, and edge detection, etc. Unlike CNN, the number of hidden layers and kernels applied to each layer is limited in MNN. Because of the reduction of processing time and utility of VLSI chip design, it is possible to apply MNN to various mobile embedded systems. MNN performs the edge and shape detection operations with a limited number of kernels. Through experiments using database images, it is confirmed that MNN can be used as a deep learning system and its performance.

• The Journal of the Korea Contents Association
• /
• v.8 no.5
• /
• pp.276-286
• /
• 2008

In this paper, we introduce an adaptive e-Learning system for engineering mathematics which is based on computer algebra system (Mathematica) and on-line authoring environment. The system provides an assessment tool for individual diagnosis using Bayesian inference network. Using this system, an instructor can easily develop mathematical web contents via web interface. Examples of such content development are illustrated in the area of linear algebra, differential equation and discrete mathematics. The diagnostic module traces a student's knowledge level based on statistical inference using the conditional probability and Bayesian updating algorithm via Netica. As part of formative evaluation, we brought this system into real university settings and analyzed students' feedback using survey.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2015.07a
• /
• pp.224-225
• /
• 2015

IT를 기반으로 한 플립러닝(Flipped Learning-온라인으로 선행학습을 한 뒤에 오프라인 수업을 통해 교사와 학생이 토론식으로 진행하는 수업방식) 학습법은 교수자가 아닌 학습자의 요구에 적극 부응할 수 있는 제도이다. 전공 수업뿐만 아니라 교양 수업에도 이에 대한 적극적인 도입이 필요하다. 본 논문에서는 대학 글쓰기 수업에 있어서 플립러닝 학습법을 적용한 사례를 통해 이를 입증하고자 하였다.

• Proceedings of the Korea Contents Association Conference
• /
• 2010.05a
• /
• pp.569-572
• /
• 2010

맞춤형 이러닝은 학습효과의 증진을 위한 방안으로 개인 맞춤형, 개인화 혹은 적응화 등의 개념이 제안되고 확장되었다. 본 논문에서는 초기 선호도, 흥미도 혹은 검색습관을 고려하는 방식으로부터, 특정한 학습객체를 자율 선택하여 반복 학습할 수 있도록 개념단위를 적용한 방식, 학습자의 능력을 고려한 최적 난이도 학습객체를 제공하는 방식 등 다양하게 제안되고 있는 현대적 개인 맞춤형 이러닝 체계들을 비교 분석한다. 개별 시스템에 따라 '평가'에 국한되거나 '평가'와 '학습'을 연계하는 경우가 존재하며, 이에 따른 적용에 의하여 학습환경과 맞춤형 제공방식 및 학습효과를 상호 연계할 수 있음을 밝힌다.