• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권11호
• /
• pp.2903-2923
• /
• 2023

Knowledge recommendation is a type of recommendation system that recommends knowledge content to users in order to satisfy their needs. Although using graph neural networks to extract data features is an effective method for solving the recommendation problem, there is information loss when modeling real-world problems because an edge in a graph structure can only be associated with two nodes. Because one super-edge in the hypergraph structure can be connected with several nodes and the effectiveness of knowledge graph for knowledge expression, a dual-channel hypergraph convolutional neural network model (DCHC) based on hypergraph structure and knowledge graph is proposed. The model divides user data and knowledge data into user subhypergraph and knowledge subhypergraph, respectively, and extracts user data features by dual-channel hypergraph convolution and knowledge data features by combining with knowledge graph technology, and finally generates recommendation results based on the obtained user embedding and knowledge embedding. The performance of DCHC model is higher than the comparative model under AUC and F1 evaluation indicators, comparative experiments with the baseline also demonstrate the validity of DCHC model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권5호
• /
• pp.1814-1828
• /
• 2021

Low-light image enhancement is a key technique to overcome the quality degradation of photos taken under scotopic vision illumination conditions. The degradation includes low brightness, low contrast, and outstanding noise, which would seriously affect the vision of the human eye recognition ability and subsequent image processing. In this paper, we propose an approach based on deep learning and Retinex theory to enhance the low-light image, which includes image decomposition, illumination prediction, image reconstruction, and image optimization. The first three parts can reconstruct the enhanced image that suffers from low-resolution. To reduce the noise of the enhanced image and improve the image quality, a super-resolution algorithm based on the Laplacian pyramid network is introduced to optimize the image. The Laplacian pyramid network can improve the resolution of the enhanced image through multiple feature extraction and deconvolution operations. Furthermore, a combination loss function is explored in the network training stage to improve the efficiency of the algorithm. Extensive experiments and comprehensive evaluations demonstrate the strength of the proposed method, the result is closer to the real-world scene in lightness, color, and details. Besides, experiments also demonstrate that the proposed method with the single low-light image can achieve the same effect as multi-exposure image fusion algorithm and no ghost is introduced.

• International journal of advanced smart convergence
• /
• 제4권2호
• /
• pp.38-45
• /
• 2015

Peer-to-Peer(P2P) systems are generally classified into two categories; hybrid and pure P2P. Hybrid systems have a single central index server keeping the details of shared information, so that undesirable effects such as heavy load on that server and lack of fault-tolerance can be caused. Pure P2P causes the other problems such as message flooding and scalability although it shows high degree of fault-tolerance. Recently, mobile agent-based distributed computing has been receiving wide attention for its potential to support disconnected operations, high asynchrony, and thus saving network bandwidth. In this paper, a new scheme of peer coordination is proposed for a decentralized P2P network with self-organizing structure. We deployed mobile agents for incorporating the advantages of usage of mobile agents into our P2P network. Proposed P2P network has both advantages of hybrid and pure P2P. The problems of heavy load on the server and lack of fault-tolerance are improved by using multiple special peers called super-peers. And the problems of pure P2P can be reduced by using mobile agents.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권5호
• /
• pp.2103-2123
• /
• 2018

Ships built today are larger in scale and feature more complex structures. The ever-evolving systems used on board a ship require vast amounts of data processing. In the future, with the advent of smart ships, unmanned ships and other next-generation ships, the volumes of data to be processed will continue to increase. Yet, to date, ship data has been processed using wired networks. Placed at fixed locations, the nodes on wired networks often fail to process data from mobile devices. Despite many attempts made to use Wi-Fi on ships just as on land to create wireless networks, Wi-Fi has hardly been available due to the complex metal structures of ships. Therefore, Wi-Fi on ships has been patchy as the ship-wide total Wi-Fi coverage has not properly implemented. A new ship-wide wireless network environment is part of the technology conducive to the shipbuilding industry. The wireless network environment should not only serve the purpose of communication but also be able to manage and control multiple features in real-time: fault diagnostics, tracking, accident prevention and safety management. To better understand the characteristics of wireless frequencies for ships, this paper tests the widely used TETRA, UHF and Wi-Fi and sheds light on the features, advantages and disadvantages of each technology in ship settings. The proposed deployment of a Super Wi-Fi network leveraging the legacy UHF system of TMS generates a ship-wide wireless network environment. The experimental findings corroborate the feasibility of the proposed ship-wide Super Wi-Fi network environment.

• 방송공학회논문지
• /
• 제25권5호
• /
• pp.672-684
• /
• 2020

시차 기반 영상처리에 대한 연구들이 증가함에 따라 저해상도 및 모션 블러된 라이트필드 영상을 복원하는 연구는 필수적이 되었다. 이러한 기법들은 라이트필드 영상 향상 과정으로 알려져 있으나 두 개 이상의 문제를 동시에 해결하는 기존의 연구는 거의 존재하지 않는다. 본 논문에서는 라이트필드 공간 영역 초해상도 복원과 모션 블러 제거를 동시 수행하는 프레임워크를 제안한다. 특히, 저해상도 및 6-DOF 모션 블러된 라이트필드 데이터셋으로 훈련하는 간단한 네트워크를 생성한다. 또한 성능을 향상하기 위해 생성적 적대 신경망의 지역 영역 최적화 기법을 제안하였다. 제안한 프레임워크는 정량적, 정성적 측정을 통해 평가하고 기존의 state-of-the-art 기법들과 비교하여 우수한 성능을 나타냄을 보인다.

• 한국정보통신학회논문지
• /
• 제5권2호
• /
• pp.239-245
• /
• 2001

모든 국민에 대한 기본통신의 보편적 서비스 제공 정책에 힘입어 일반 공중전화망에 의한 기본 통신 서비스 부문에서의 지역 간의 격차는 상당히 해소되었으나, 최근 정보화가 급격히 추진됨에 따라 고도 정보통신분야의 서비스에 있어서는 지역별로 편중현상이 발생하고 있다. 특히, 도서지역은 환해성과 격리성 등의 지리적 요인 및 인구밀도, 소득 및 통화량 등이 육지에 비해 현저히 낮은 관계로 가입자망 고도화가 이루어지지 않고 있다. 이에 따라 도서지역은 정보화의 소외 지역으로 낡게되어 지역 간의 발전격차가 심화됨은 물론 다가오는 해양시대를 능동적으로 대처하는데 있어서 커다란 문제점을 안고 있다. 본 논문은 현재 우리나라 도서지역 통신망의 구성 운용현황 및 문제점 등을 분석하고 도서 주민에게 고도 정보서비스를 경제적, 효율적으로 제공하기 위한 가입자 망 고도화 방안에 대해 검토하였다.

• 디지털융복합연구
• /
• 제11권4호
• /
• pp.215-220
• /
• 2013

슈퍼 컴퓨터는 20G 프리플롭을 처리 할 수 있는 용량을 가진 컴퓨터를 말하며, 과거에는 슈퍼 컴퓨터로 복잡한 과학 계산을 처리하였으며, 지금도 많은 클라이언트 컴퓨터가 복잡한 계산 처리를 요청하면 빠른 처리 속도로 클라이언트의 요청을 처리하고 있다. 슈퍼 컴퓨터가 바이러스에 감염될 경우 연결된 많은 클라이언트 컴퓨터에 영향을 미칠 수 있기 때문에 최근 사이버 공격은 슈퍼 컴퓨터에 초점을 맞추고 있고, 따라서, 본 연구 논문은 슈퍼 컴퓨터 보안 문제를 분석하고 슈퍼 컴퓨터의 보안 레벨을 올리기 위해 현재 문제점과 생체정보를 이용한 방어책을 제시하였다.

• Journal of information and communication convergence engineering
• /
• 제9권4호
• /
• pp.396-400
• /
• 2011

P2P(peer to Peer) network is a distributed network architecture composed of participants that make a portion of their resources directly available to other network participants, without the need for a central server. Currently, convergence network industry using wired and mobile are grows rapidly. So P2P protocols will be used between mobile and wired network. But current P2P protocols are focused on the wired networks only and there are no simulators for performance analysis of mobile P2P. In this paper, we design a P2P simulation system for performance analysis of P2P protocols in mobile, wired and convergence networks. It is constructed by a well-known mobile network simulator and wired based P2P protocol simulator. Finally we have implemented a smart TV test-bed using our P2P test-bed for convergence networks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권10호
• /
• pp.4968-4986
• /
• 2017

In this paper, we primarily address the difficulty of automatic generation of a plausible depth map from a single image in an unstructured environment. The aim is to extrapolate a depth map with a more correct, rich, and distinct depth order, which is both quantitatively accurate as well as visually pleasing. Our technique, which is fundamentally based on a preexisting DepthTransfer algorithm, transfers depth information at the level of superpixels. This occurs within a framework that replaces a pixel basis with one of instance-based learning. A vital superpixels feature enhancing matching precision is posterior incorporation of predictive semantic labels into the depth extraction procedure. Finally, a modified Cross Bilateral Filter is leveraged to augment the final depth field. For training and evaluation, experiments were conducted using the Make3D Range Image Dataset and vividly demonstrate that this depth estimation method outperforms state-of-the-art methods for the correlation coefficient metric, mean log10 error and root mean squared error, and achieves comparable performance for the average relative error metric in both efficacy and computational efficiency. This approach can be utilized to automatically convert 2D images into stereo for 3D visualization, producing anaglyph images that are visually superior in realism and simultaneously more immersive.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2004년도 ICEIC The International Conference on Electronics Informations and Communications
• /
• pp.193-198
• /
• 2004

In variety of environments for applications, wireless sensor networks have received increasing attention in the recent few years. But, sensor nodes have many limitations including battery power and communication range. These networks require robust wireless communicant protocols that are energy efficient and provide low latency. In this paper, we propose new protocol as is defined SCP. The key idea of SCP is that only one node which is defined as a Super-Cluster Header sends the combined data to the BS. We evaluated the effectiveness of SCP through experiments which have several parameter violations. Simulation results shows that performance of SCP is through better than other legacy protocol within the framework of energy cost, life time of the sensor network and fair distribution of the energy consumption.