• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.292-293
• /
• 2018

최근 증강현실 산업 분야가 많은 각광을 받고, 시장이 성장함에 따라 보다 쉽게 증강현실을 구현 할 수 있도록 많은 SDK(Software Development Kit)들이 발표되었다. 기존에 발표 된 SDK들이 대부분 강체 추적만을 제공한다. 이는 현재 추적 알고리즘의 기반이 되는 이론이 강체에 한정되어 있기 때문이다. 그러나 제안하는 프레임워크는 강체 추적뿐만 아니라 비강체 추적 또한 가능하다. 이를 위하여, 제안하는 프레임워크는 증강현실의 핵심 기술인 추적 엔진과 보다 넓은 확장성을 가지도록 추적하고자 하는 물체를 사전에 분석하고 실시간으로 모델 변형 정보를 추정하는 시뮬레이션 엔진으로 구성된다. 추적 엔진은 기본적으로 물체의 표면에 존재하는 특징점 정보를 이용하여 추적을 진행 하되, 비강체 추적을 위하여 시뮬레이션 엔진의 도움을 받는 형태로 구성된다. 시뮬레이션 엔진에서는 물체의 역학 파라미터를 추정하여 이를 추적을 진행 할 때, 추적 엔진의 물체 표면 특징점 정보를 이용하여 물체의 변형 정보를 추정한다. 또한 제안하는 프레임워크는 성능 상의 장점 외에도 오픈소스로 공개되기에 국내 증강현실 시장 성장에 발판이 될 것으로 기대된다.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.7
• /
• pp.313-318
• /
• 2016

To utilize sensory data, it is necessary to design architecture for processing and handling data generated from sensors in an IoT environment. Especially in the IoT environment, a thing connects to the Internet and efficiently enables to communicate a device with diverse sensors. But Hadoop and Twister based on MapReduce are good at handling data in a batch processing. It has a limitation for processing stream data from a sensor in a motion. Traditional streaming data processing has been mainly applied a MoM based message queuing system. It has maintainability and scalability problems because a programmer should consider details related with complex messaging flow. In this paper architecture is designed to handle sensory data aggregated The designed software architecture is used to operate an application on the open source framework Storm. The application is conceptually used to transform streaming data which aggregated via sensor gateway by pipe-filter style.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2018.01a
• /
• pp.75-76
• /
• 2018

본 논문에서는 상품증권 거래 시스템에 블록체인을 적용하여, 여러 서비스에 거래 시스템을 적용할 수 있게 설계하였다. 블록체인 오픈소스 프레임워크인 Hyperledger를 사용하여 시스템의 안정성을 높였다. 단말기의 배터리 소모량과 네트워크 트래픽을 감소시키기 위한 방식으로 거래소를 적용하는 방법에 관해 기술하였다. 또한, 서드 파티 개발자가 플랫폼을 쉽게 구축할 수 있도록 API 구조를 설계하였다.

• Proceedings of the KSRS Conference
• /
• 2009.03a
• /
• pp.219-219
• /
• 2009

• The Journal of the Korea Contents Association
• /
• v.9 no.2
• /
• pp.133-141
• /
• 2009

This paper propose an expanded ASP exclusive development framework method. We applied a module to the suggested method which suit to develope ASP. Proposed method is a kind of module, which manages MVC develope methodology and treats Query and ASP information as a XML. It has flexibility which meets the client's request quickly. Using inherited base model which previously exist in template source, developer only concerned with development methodology so that they can reduce development time and reinforce a integrity of system. We proposed these improvements as a framework. Experimental results validate our scheme, showing a fewer number of files and a better flexibility than other development method, and also about 29% of program code has been reduced.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.11
• /
• pp.2593-2599
• /
• 2014

The way of mobile apps development using open source software have been becoming increasingly popular recently. In opensource mobile User Interface components case, there is tends to much more be used than mobile functional components. Because it is more easier that the reusability for implementation of User Interface. The problem is to apply as an open source mobile components selected for this reason all of the existing two studies. It is an open source software selection process and selection procedures shelf components. We revise to conform to existing mobile components selected for the study on open source software selection process in this paper. Can help increase the productivity of the mobile app development by the open source components to meet the functional requirements when developing mobile apps that can be easily retrieved and presented the selection process of this study is effective.

• Smart Media Journal
• /
• v.7 no.2
• /
• pp.34-39
• /
• 2018

There are many studies on red tide due to the continuous increase in damage to domestic fish and shell farms by the harmful red tide. However, there is insufficient domestic research of identifying harmful red tide algae that automatically recognizes red tide images. In this paper, we propose a red tide image classification method using deep learning based open source. To solve the problem of recognition of various images of red tide algae, the proposed method is implemented by using tensorflow framework and Google image classification model.

• Communications of the Korean Institute of Information Scientists and Engineers
• /
• v.31 no.10
• /
• pp.47-53
• /
• 2013

• Journal of the Korean Society for information Management
• /
• v.34 no.1
• /
• pp.341-371
• /
• 2017

Recently, the web service environment has changed from document-centered to data-oriented focus, and the Linked Open Data (LOD) exists at the core of the new environment. Specific procedures and methods were examined to build the LOD of records information in accordance with this trend. With the service sustainability of small-scale archive in consideration, an exemplification on LOD building process by utilizing open source software was developed in this paper. To this end, a 5-step service framework for LOD construction was proposed and applied to a collection of diary records from 'Human and Memory Archive'. Proof of Concept (POC) utilizing open source softwares, Protege and Apache Jena Fuseki, was conducted according to the proposed 5 step framework. After establishing the LOD of record information by utilizing the open source software, the connection with external LOD through interlinking and SPARQL search has been successfully performed. In addition, archives' considerations for LOD construction, including improvement on the quality of content information, the role of the archivist, were suggested based on the understanding obtained through the LOD construction process of records information.

• KIISE Transactions on Computing Practices
• /
• v.22 no.11
• /
• pp.547-558
• /
• 2016

Construction of multi-site Cloud Infrastructure with OpenStack, the most popular cloud open source project with various ICT infrastructure technologies, has operational inefficiency due to installation complexity and geographical limitation. To address this inefficiency, we gradually developed automated installation/configuration tools to automate installation/configuration of Linux and OpenStack by complying with DevOps methodology over a multi-site OpenStack testbed called OF@KOREN SmartX Playground. However, the pre-developed installation tools can be used only for limited cases. Therefore, we designed and developed a prototype of SmartX Provisioning Framework which could conducts Playground-wide provisioning flexibly by following three steps of resource management in Software-defined Infrastructure. We verified the efficiency of the provisioning functionality of the framework by demonstrating an example of automated multi-site cloud construction.