• Proceedings of the Technology Innovation Conference
• /
• 2005.02a
• /
• pp.268-286
• /
• 2005

본 연구에서는 OECD 주요 10개국을 대상으로 국가별 정보통신산업의 성장 추이를 각각 분석하고 국별 특성을 비교하는데 목적이 있다. 이를 바탕으로 각국의 정보통신산업이 경기순환 또는 단계별 발전 속성을 지니고 있는지를 파악하고 국가별 공통점과 특이점을 분석하고자 하였다. 방법론적으로 OECD 국가들의 정보통신산업 GDP 추이 및 성장률의 움직임을 국면전환 (regime change) 확산과정으로 묘사함으로써 각 국가별 정보통신산업 발전 양상의 특징 및 국면전환 시점 등을 포착해 내고자 하였다 추세를 갖는 대표적 확산과정인 GBM 모형과 평균회귀 성향을 갖는 대표적 확산과정인 Vasicek 모형에 각각 마코프 국면전환을 도입하여 국가별 정보통신산업 GDP 및 GDP 성장률의 추이에 있어 국면 전환 여부와 독특한 발전 특성을 비교 분석하였다. 실증분석 결과 정보통신산업 GDP의 성장률과 변동성 사이에는 높은 상관관계가 있었으며, 한국, 멕시코 등은 고성장, 고변동성을, 미국, 프랑스, 일본 등은 저성장, 저변동성의 특성을 보이는 것으로 나타났다 또한 한국의 경우 유일하게 성장률과 변동성 모두 국면전환이 일어나는 국가로 나타났다. 장기평균 성장률의 특성에 따라 분류한 결과, 한국, 일본, 미국, 멕시코, 뉴질랜드는 고성장에서 저성장으로의 국면전환, 핀란드와 덴마크는 경기 순환적 국면전환, 노르웨이, 프랑스, 캐나다는 단일 국면으로 분류할 수 있었다. 특히 한국의 경우 평균회귀 속도와 변동성이 타 국가에 비해 높은 특성을 보여주었다. 본 연구는 정보통신산업을 미시적 분석이나 세부 항목별 정량적 분석을 통해서가 아니라 산업의 발전 속성 및 경기 순환 등의 관점에서 분석함으로써 정보통신산업 정책의 수립 및 집행을 거시적 안목 하에 정립할 수 있게 한다는 데 의의를 가진다. 또한 경제변수를 묘사하는데 있어 국면전환 확산과정을 사용함으로써 향후 실물옵션 등을 통한 기술 및 무형자산의 가치평가에 있어 기초자산의 움직임을 보다 정확히 포착해 낼 수 있는 프로세스를 제공하였다는데 또 다른 의의를 갖는다고 하겠다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.7
• /
• pp.1267-1274
• /
• 2007

In this paper, we propose a digital watermarking algorithm for digital hologram generated by computer generated hologram(CGH) method. Digital hologram generated by a computer calculation(CGH) is one of the most expensive contents and its usage is being expanded. Thus, it is highly necessary to protect the ownership of digital hologram. In this paper two digital watermarking schemes are introduced hologram-domain and global 2DDCT-domain scheme. Proposed watermarking scheme showed very high imperceptibility and quite high robustness against the attacks. The purpose of this paper is to introduce these global 2DDCT based watermarking schemes. Thus, we expect that these and the contents in this paper can be very useful bases for the further digital watermarking schemes of the digital holograms.

• Journal of Software Assessment and Valuation
• /
• v.17 no.1
• /
• pp.51-58
• /
• 2021

Microservices are not only developed independently, but can also be run and deployed independently, ensuring more flexible scaling and efficient collaboration in a cloud computing environment. This impact has led to a surge in migrating to microservices-oriented application environments in recent years. In order to introduce microservices, the problem of identifying microservice units in a single application built with a single architecture must first be solved. In this paper, we propose an algorithm-based approach to identify microservices from legacy systems. A graph is generated using the meta-information of the legacy code, and a microservice candidate is extracted by applying a clustering algorithm. Modularization quality is evaluated using metrics for the extracted microservice candidates. In addition, in order to validate the proposed method, candidate services are derived using codes of open software that are widely used for benchmarking, and the level of modularity is evaluated using metrics. It can be identified as a smaller unit of microservice, and as a result, the module quality has improved.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.448-453
• /
• 2023

대규모 언어 모델 (Large Language Model, LLM)을 인간의 선호도 관점에서 평가하는 것은 기존의 벤치마크 평가와는 다른 도전적인 과제이다. 이를 위해, 기존 연구들은 강력한 LLM을 평가자로 사용하여 접근하였지만, 높은 비용 문제가 부각되었다. 또한, 평가자로서 LLM이 사용하는 주관적인 점수 기준은 모호하여 평가 결과의 신뢰성을 저해하며, 단일 모델에 의한 평가 결과는 편향될 가능성이 있다. 본 논문에서는 엄격한 기준을 활용하여 편향되지 않은 평가를 수행할 수 있는 평가 프레임워크 및 평가자 모델 'FubaoLM'을 제안한다. 우리의 평가 프레임워크는 심층적인 평가 기준을 통해 다수의 강력한 한국어 LLM을 활용하여 연쇄적 사고(Chain-of-Thought) 기반 평가를 수행한다. 이러한 평가 결과를 다수결로 통합하여 편향되지 않은 평가 결과를 도출하며, 지시 조정 (instruction tuning)을 통해 FubaoLM은 다수의 LLM으로 부터 평가 지식을 증류받는다. 더 나아가 본 논문에서는 전문가 기반 평가 데이터셋을 구축하여 FubaoLM 효과성을 입증한다. 우리의 실험에서 앙상블된 FubaoLM은 GPT-3.5 대비 16% 에서 23% 향상된 절대 평가 성능을 가지며, 이항 평가에서 인간과 유사한 선호도 평가 결과를 도출한다. 이를 통해 FubaoLM은 비교적 적은 비용으로도 높은 신뢰성을 유지하며, 편향되지 않은 평가를 수행할 수 있음을 보인다.

• Smart Media Journal
• /
• v.12 no.9
• /
• pp.103-115
• /
• 2023

A Korean Tea-Historical Literature DB was constructed, collating information related to tea from eight historical texts, including the Annals of the Joseon Dynasty. Data were gathered from existing historical literature DB and web services. After processing, verification, and tag input, 53,679 historical literature data entries related to tea were created. This DB construction centralizes previously dispersed reading services into a single web service. Features such as literature title, time, and tag-based search functionality; assignment of attribute tags to data; search linkage through original source links; and a bookmark function were incorporated to enhance user convenience. The outcomes of this research support studies in the related fields, and it is believed that by accumulating and improving accessibility to tea-related literature, contributions can be made to the preservation and education of Korean traditional tea as a food resource and cultural heritage.

• Journal of the Korea Computer Graphics Society
• /
• v.15 no.4
• /
• pp.1-11
• /
• 2009

In this paper, we present a prism-based mesh culling method to improve effectiveness of continuous collision detection which is a major bottleneck in a simulation using polygonal mesh models. A prism is defined based on two matching triangles between a sequence of times m a polygonal model. In order to detect potential colliding set(PCS) of prism between two polygonal models in a unit time, we apply the visibility test based on the occlusion query to two sets of prisms which are defined from two polygonal models in a unit time. Moreover, we execute the narrow band culling based on SAT(Separating Axis Test) to define potential colliding prism pairs from PCS of prisms extracted as a result of the visibility test. In the SAT, we examine one axis to be perpendicular to a plane which divides a 3D space into two half spaces to include each prism. In the experiments, we applied the proposed culling method to pairs of polygonal models with the different size and compared the number of potential colliding prism pairs with the number of all possible prism pairs of two polygonal models. We also compared effectiveness and performance of the visibility test-based method with those of the SAT-based method as the second narrow band culling. In an experiment using two models to consist of 2916 and 2731 polygons, respectively, we got potential colliding prism pairs with 99 % of culling rate.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.21 no.6
• /
• pp.183-190
• /
• 2011

In this paper, we propose a prototype system for abnormal sound detection and identification which detects and recognizes the abnormal situations by means of analyzing audio information coming in real time from CCTV cameras under surveillance environment. The proposed system is composed of two layers: The first layer is an one-class support vector machine, i.e., support vector data description (SVDD) that performs rapid detection of abnormal situations and alerts to the manager. The second layer classifies the detected abnormal sound into predefined class such as 'gun', 'scream', 'siren', 'crash', 'bomb' via a sparse representation classifier (SRC) to cope with emergency situations. The proposed system is designed in a hierarchical manner via a mixture of SVDD and SRC, which has desired characteristics as follows: 1) By fast detecting abnormal sound using SVDD trained with only normal sound, it does not perform the unnecessary classification for normal sound. 2) It ensures a reliable system performance via a SRC that has been successfully applied in the field of face recognition. 3) With the intrinsic incremental learning capability of SRC, it can actively adapt itself to the change of a sound database. The experimental results with the qualitative analysis illustrate the efficiency of the proposed method.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.6
• /
• pp.1321-1330
• /
• 2023

This study attempts to address the problem of 3D pose estimation for multiple human objects through a single image generated during the character development process that can be used in augmented reality. In the existing top-down method, all objects in the image are first detected, and then each is reconstructed independently. The problem is that inconsistent results may occur due to overlap or depth order mismatch between the reconstructed objects. The goal of this study is to solve these problems and develop a single network that provides consistent 3D reconstruction of all humans in a scene. Integrating a human body model based on the SMPL parametric system into a top-down framework became an important choice. Through this, two types of collision loss based on distance field and loss that considers depth order were introduced. The first loss prevents overlap between reconstructed people, and the second loss adjusts the depth ordering of people to render occlusion inference and annotated instance segmentation consistently. This method allows depth information to be provided to the network without explicit 3D annotation of the image. Experimental results show that this study's methodology performs better than existing methods on standard 3D pose benchmarks, and the proposed losses enable more consistent reconstruction from natural images.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.8C
• /
• pp.1113-1124
• /
• 2004

In this paper, we proposed a hardware(H/W) structure which can compress and recontruct the input image in real time operation and implemented it into a FPGA platform using VHDL(VHSIC Hardware Description Language). All the image processing element to process both compression and reconstruction in a FPGA were considered each of them was mapped into H/W with the efficient structure for FPGA. We used the DWT(discrete wavelet transform) which transforms the data from spatial domain to the frequency domain, because use considered the motion JPEG2000 as the application. The implemented H/W is separated to both the data path part and the control part. The data path part consisted of the image processing blocks and the data processing blocks. The image processing blocks consisted of the DWT Kernel fur the filtering by DWT, Quantizer/Huffman Encoder, Inverse Adder/Buffer for adding the low frequency coefficient to the high frequency one in the inverse DWT operation, and Huffman Decoder. Also there existed the interface blocks for communicating with the external application environments and the timing blocks for buffering between the internal blocks The global operations of the designed H/W are the image compression and the reconstruction, and it is operated by the unit of a field synchronized with the A/D converter. The implemented H/W used the 69%(16980) LAB(Logic Array Block) and 9%(28352) ESB(Embedded System Block) in the APEX20KC EP20K600CB652-7 FPGA chip of ALTERA, and stably operated in the 70MHz clock frequency. So we verified the real time operation of 60 fields/sec(30 frames/sec).

• Journal of Soil and Groundwater Environment
• /
• v.10 no.5
• /
• pp.1-10
• /
• 2005

Policies of the United States, the United Kingdom, the Netherlands, Germany and Denmark were compared and analyzed on risk assessment of contaminated sites. These countries were chosen from a feasible preliminary analysis of 18 countries of the European Union and the U. S. All the countries selected met two major criteria : I) implementation of risk assessment to determine the soil contamination and remediation targets of contaminated sites, ii) use of soil guidance values and risk assessment as complementary measures to determine soil contamination. Suggested policy improvements to Korea regarding these issues include i) legislation of a rational risk assessment methodology of contaminated sites, and ii) enactment of collaboration of risk assessment with the soil guidance values. To establish effective risk assessment legislation, additional in-depth research on social, economic and long-term effects of the proposed risk assessment methodologies, as well as the mutual consent of all parties including academia, industry, and administration will be necessary. Linking risk assessment with soil guidance values would be applicable to a site contaminated where the contaminant concentration exceeds a certain soil guidance value. In parallel, application of risk assessment to a site where a contaminant concentration is naturally different such as mining sites would be plausible. The policy suggestions above are not yet conclusive due to a lack of policy implementation, and simulation. Thus, additional research on developing risk assessment methodology is needed. Nevertheless, initiation of the suggested policy would increase the efficacy of Korean policy regarding the survey and remediation of contaminated sites.