• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.01a
• /
• pp.447-448
• /
• 2023

본 연구에서는 비전공자들을 위한 디지털 교육을 위하여 노 코드 프로그램을 활용한 정보, 수학, 과학 융합교육 교양 강좌를 개발하였다. 노 코드 프로그램으로는 오렌지3 데이터 마이닝을 선정하였는데, 이는 데이터 분석, 시각화, 머신러닝 모델의 활용이 용이하다는 강점을 가지고 있다. 또한, 산업환경 변화에 대비하는 핵심 교과인 과학, 수학, 정보의 중요성과 데이터 분석과의 밀접성을 고려하여 교육 내용을 융합할 수 있도록 선정하였다. 개발된 교육 프로그램은 8인이 전문가 검토 결과 내용 타당도가 확보되었음을 확인할 수 있었다. 추후 연구에서는 이 강좌를 대학의 학부생에게 적용하여 그 효과성을 확인해 보고자 한다.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.12
• /
• pp.1109-1114
• /
• 2015

This paper presents an improved learning-based visual inspection method for auto parts inspection in severe lighting changes. Automobile sunroof frames are produced automatically by robots in most production lines. In the sunroof frame manufacturing process, there is a quality problem with some parts such as volts are missed. Instead of manual sampling inspection using some mechanical jig instruments, a learning-based machine vision system was proposed in the previous research[1]. But, in applying the actual sunroof frame production process, the inspection accuracy of the proposed vision system is much lowered because of severe illumination changes. In order to overcome this capricious environment, some selective feature vectors and cascade classifiers are used for each auto parts. And we are able to improve the inspection accuracy through the re-learning concept for the misclassified data. The effectiveness of the proposed visual inspection method is verified through sufficient experiments in a real sunroof production line.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.7
• /
• pp.621-626
• /
• 2015

This paper presents a camera calibration method in order to estimate the lane detection and inter-vehicle distance estimation system for an automotive safety driving system. In order to implement the lane detection and vision-based inter-vehicle distance estimation to the embedded navigations or black box systems, it is necessary to consider the computation time and algorithm complexity. The process of camera calibration estimates the horizon, the position of the car's hood and the lane width for extraction of region of interest (ROI) from input image sequences. The precision of the calibration method is very important to the lane detection and inter-vehicle distance estimation. The proposed calibration method consists of three main steps: 1) horizon area determination; 2) estimation of the car's hood area; and 3) estimation of initial lane width. Various experimental results show the effectiveness of the proposed method.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.11a
• /
• pp.1119-1121
• /
• 2014

저널 논문 투고 및 심사시스템에서의 논문 제출은 상시 이루어진다는 특성 때문에 논문이 제출된 시점에 적절한 심사자들을 찾아 배정하기란 쉽지 않은 문제이다. 본 논문에서는 이러한 문제를 해결하기 위하여 제출된 논문에 적절한 심사자들을 추천해주는 알고리즘을 제시하고자 한다. 심사자 추천 알고리즘에서는 해당 논문의 전문가를 심사자로써 추천하기 위하여 제출된 논문들의 키워드(Keyword)와 심사자들의 전문지식태그(Expertise Tag) 정보를 활용한다. 또한 심사자들의 기존의 심사 정보를 토대로 심사활동지수를 평가하여 이를 심사자 추천에 활용하고자 한다. 제안하는 알고리즘을 검증하기 위하여 본 논문에서는 실제 저널 논문투고시스템에 추천 알고리즘을 적용해보고 이의 결과를 제시한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.11a
• /
• pp.1126-1129
• /
• 2014

사회관계망 분석에 있어서 매개중심도는 네트워크를 구성하는 노드들의 상대적인 중요도를 파악하기 위한 척도로서 오랫동안 사용되어 왔다. 그러나 이러한 매개중심도를 계산하기 위한 계산 복잡도가 높기 때문에 대규모 사회관계망에서는 매개중심도를 계산하기란 쉽지 않은 문제이다. 본 논문에서는 네트워크를 구성하는 각각의 노드들마다 자신의 지역정보를 활용하여 구성한 네트워크에서 매개중심도를 산출함으로써 시간복잡도를 줄이는 한편 지역정보 기반의 네트워크의 특징을 분석함으로써 매개중심도를 더 빠르게 산출할 수 있는 알고리즘을 제안한다. 그리고 실제 소셜 네트워크에서의 실험을 통하여 제안 알고리즘이 기존 알고리즘에 비해 매개중심도를 더 빠르게 산출함을 보인다.

• Journal of Broadcast Engineering
• /
• v.27 no.5
• /
• pp.773-793
• /
• 2022

The moving picture experts group (MPEG) immersive video (MIV) coding standard has been established to support six degrees of freedom (6DoF) in virtual reality (VR) by transmitting high-quality multiple immersive videos. The MIV exploits two approaches considering tradeoff between bandwidth and computational complexity: 1) eliminating correlation between multi-view videos or 2) selecting representative videos. This paper presents performance analysis on intermediate synthesized views on source view positions and synthesized pose traces using high-efficiency video coding (HEVC) and versatile video coding (VVC) for above-mentioned two approaches.

• Journal of information and communication convergence engineering
• /
• v.20 no.4
• /
• pp.288-294
• /
• 2022

The media content market has been growing, as various types of content are being mass-produced owing to the recent proliferation of the Internet and digital media. In addition, platforms that provide personalized services for content consumption are emerging and competing with each other to recommend personalized content. Existing platforms use a method in which a user directly inputs video metadata. Consequently, significant amounts of time and cost are consumed in processing large amounts of data. In this study, keyframes and audio spectra based on the YCbCr color model of a movie trailer were extracted for the automatic generation of metadata. The extracted audio spectra and image keyframes were used as learning data for genre recognition in deep learning. Deep learning was implemented to determine genres among the video metadata, and suggestions for utilization were proposed. A system that can automatically generate metadata established through the results of this study will be helpful for studying recommendation systems for media super-personalization.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.253-256
• /
• 2022

실사 영상 기반의 메타버스 환경을 구축하기 위한 다수의 카메라를 통한 영상 취득 및 부호화, 전송 기술이 활발히 연구되고 있고, 이를 위해 영상 압축 표준화 단체인 moving picture experts group (MPEG) 에서는 MPEG immersive video (MIV) 표준을 개발하였다. 하지만, 현재 널리 사용되는 가상 현실 영상을 스트리밍 가능한 장비의 연산 능력으로는 MIV 기반 몰입형 영상을 스트리밍 시 복호기 동기화 문제가 발생할 수 있다. 따라서 본 논문은 저사양 및 고사양 장비에서 적응적으로 복호기 개수를 조절 가능한 geometry packing 기법을 MIV 의 참조 소프트웨어인 test model for immersive video (TMIV)에서 구현한다. 제안하는 패킹 기법은 지오메트리 영상을 패킹하여 텍스쳐 영상과 같은 높이를 가지도록 한 후 각각 단일 서브픽쳐 (subpicture) 로 부호화한다. 이후 부호화된 서브픽쳐들에 적응적으로 비트스트림 병합이 적용되어 장비의 복호기 사양에 대응한다.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.4
• /
• pp.99-102
• /
• 2023

In this paper, the system is proposed and implemented to share the part number, the part name, and the vehicle type through the improvement sharing bulletin board for automobile repair and maintenance. And when photos of damage parts are uploaded to the system, the system analyzes it using a deep learning model to analyze whether it is damaged and automatically classify the type of damage. By providing repair and maintenance quotes for a significant part, the system provides economically repaired by providing comparative adjustment information on repair costs to drivers who are particularly concerned about the market prices of parts and maintenance services. Through the existing bulletin board, you can exchange and share information about parts by sharing various information on repair and maintenance. This paper provides in detail the average market price per type of damage during automobile repair and maintenance, helping drivers who do not know the details of parts and maintenance services to receive reasonable quotes by providing price information.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3416-3435
• /
• 2023

Task offloading in vehicular networks is hot topic in the development of autonomous driving. In these scenarios, due to the role of vehicles and pedestrians, task characteristics are changing constantly. The classical deep learning algorithm always uses a pre-trained neural network to optimize task offloading, which leads to system performance degradation. Therefore, this paper proposes a neural network fine-tuning task offloading algorithm, combining with location prediction for pedestrians and vehicles by the Payne model of fluid dynamics and the car-following model, respectively. After the locations are predicted, characteristics of tasks can be obtained and the neural network will be fine-tuned. Finally, the proposed algorithm continuously predicts task characteristics and fine-tunes a neural network to maintain high system performance and meet low delay requirements. From the simulation results, compared with other algorithms, the proposed algorithm still guarantees a lower task offloading delay, especially when congestion occurs.