• Journal of KIISE:Software and Applications
• /
• v.35 no.12
• /
• pp.753-763
• /
• 2008

The amount of UCCs (User Created Contents) has been increasing rapidly and is associated with a serious copyright problem. Automatic logo detection in videos is an efficient means of overcoming the copyright problem. However, logos have varying characteristics, which make logo detection and recognition very difficult. Especially, there are frequent logo transitions in a video, comprising several video contents. This disrupts accurate video segmentation based on logos. Therefore, this paper proposes an accurate logo transition detection method for recognizing logos in digital video contents. The proposed method accurately segments a video according to logo and efficiently recognizes various types of logos. The experimental results demonstrate the effectiveness of the proposed method for logo detection and video segmentation according to logo.

• Proceedings of the KSRS Conference
• /
• v.1
• /
• pp.409-412
• /
• 2006

Today's commercial high resolution satellite imagery such as IKONOS and QuickBird, offers the potential to extract useful spatial information for geographical database construction and GIS applications. Recognizing this potential use of high resolution satellite imagery, KARI is performing a project for developing Korea multipurpose satellite 3(KOMPSAT-3). Therefore, it is necessary to develop techniques for various GIS applications of KOMPSAT-3, using similar high resolution satellite imagery. As fundamental studies for this purpose, we focused on the extraction of 3D spatial information and the update of existing GIS data from QuickBird imagery. This paper examines the scheme for rectification of high resolution image, and suggests the convenient semi-automatic algorithm for extraction of 3D building information from a single image. The algorithm is based on triangular vector structure that consists of a building bottom point, its corresponding roof point and a shadow end point. The proposed method could increase the number of measurable building, and enhance the digitizing accuracy and the computation efficiency.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.95-100
• /
• 2018

채팅 시스템을 잘 만들기 위해서는 양질, 대량의 채팅 말뭉치가 굉장히 중요하지만 구축 시 많은 비용이 발생한다는 어려움이 있었다. 따라서 본 논문에서는 영화 자막, 극대본과 같이 대량의 발화 데이터를 이용하여 채팅 말뭉치를 반자동으로 확장하는 방법을 제안한다. 채팅 말뭉치 확장을 위해 미리 구축된 채팅 말뭉치와 유사도 기법을 이용하여 채팅 유사도를 구하고, 채팅 유사도가 실험을 통해 얻은 임계값보다 크다면 올바른 채팅쌍이라고 판단하였다. 그리고 길이가 매우 짧은 채팅성 발화의 채팅 유사도를 효과적으로 계산하기 위해 본 논문에서 제안하는 것은 형태소 단위 임베딩 벡터와 합성곱 신경망 모델을 이용하여 발화 단위 표상을 생성하는 것이다. 실험 결과 기본 발화 단위 표상 생성 방법인 TF를 이용하는 것보다 정확률, 재현율, F1에서 각각 5.16%p, 6.09%p, 5.73%p 상승하여 61.28%, 53.19%, 56.94%의 성능을 가지는 채팅 말뭉치 반자동 구축 모델을 생성할 수 있었다.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.101-105
• /
• 2018

이 논문은 가장 잘 알려진 어휘부중 하나인 워드넷의 활용 범위 확장을 위해 워드넷 신셋에 "사건구조 프레임(Event Structure Frame)"을 주석하는 연구에 관한 것이다. 워드넷을 비롯하여 현재 사용되고 있는 어휘부는 풍부한 어휘의미정보가 구조화되어 있지만, 사건구조에 관한 정보를 포함하고 있지는 않다. 이 연구의 가장 큰 기여는 워드넷에 사건구조 프레임을 추가함으로써 워드넷과의 연결만으로 핵심적인 어휘의미정보를 모두 추출할 수 있도록 해준다는 점이다. 예를 들어 텍스트 추론, 자연어처리, 멀티 모달 태스크 등은 어휘의미정보와 배경지식(상식)을 이용하여 태스크를 수행한다. 워드넷에 대한 사건구조 주석은 자동사건구조 주석 시스템인 GESL을 이용하여 워드넷 신셋에 있는 예문에 먼저 자동 주석을 하고, 오류에 대해 수동 수정을 하는 반자동 방식이다. 사전 정의된 23개의 사건구조 프레임에 따라 예문에 출현하는 타겟 동사를 분류하고, 해당 프레임과 매핑한다. 현재 이 연구는 시작 단계이며, 이 논문에서는 빈도 순위가 가장 높은 100개의 동사와 각 사건구조 프레임별 대표 동사를 포함하여 총 106개의 동사 레마에 대해 실험을 진행하였다. 그 동사들에 대한 전체 워드넷 신셋의 수는 1337개이다. 예문이 없어서 GESL이 적용될 수 없는 신셋을 제외하면 1112개 신셋이다. 이 신셋들에 대해 GESL을 적용한 결과 F-Measure는 73.5%이다. 향후 연구에서는 워드넷-사건구조 링크를 계속 업데이트하면서 딥러닝을 이용해 GESL 성능을 향상 할 수 있는 방법을 모색할 것이다.

• Journal of Welding and Joining
• /
• v.27 no.3
• /
• pp.67-72
• /
• 2009

The flux cored arc welding (FCAW) process is one of the most frequently employed and important welding process due to high productivity and excellent workability. The process is performed either as an automated process or as a semi-automatic process. In FCAW process, welding voltage has been considered as a qualitative indication of arc length. But it is necessary to let welding operators know, maintain and manage the arc length directly by estimating and displaying it. In this study, to develop arc length estimation technique, we measured a welding circuit resistance($R_sc$) and then we calculated welding circuit voltage drop($V_sc$). Also, we measured arc peak voltage($V_ap$). By subtracting $V_sc$ from $V_arc$, we can easily calculate net arc voltage drop($V_arc$). Consequently, we suggested arc length estimating equation and basic algorithm by regressive analyzing the relationship between net arc voltage drop($V_arc$) and real arc length(Larc) measured by high speed camera. Therefore, arc length can be predicted by just monitoring welding current and voltage.

• The Journal of Society for e-Business Studies
• /
• v.12 no.1
• /
• pp.135-149
• /
• 2007

There are many kinds of applications or software components to manage files in a local computer, but it is very difficult to organize personal documents in a consistent way and to search expected ones in a precise way. In this paper, we present our development of a document management and retrieval tool, which is named Ontalk. Our system provides a semi-automatic metadata generator and an ontology-based search engine for electronic documents. Ontalk can create and import various ontologies in RDFS or OWL for describing the metadata. Our system that is built upon.NET technology is easily communicated with or flexibly plugged into many different programs.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1997.04a
• /
• pp.496-500
• /
• 1997

This paper describes an automated system for analyzing the stress intensity factors(SIFs) of three-dimensional (3D) cracks. A geometry model, i.e.a solid containing one or several 3D cracks is defined. Several distributions of local node density are chosen, and then automatically superposed on one another over the geometry model by using the fuzzy knowledge processing. Nodes are generated by the bucketing method, and ten-noded quadratic tetrahedral solid elements are generated by the Delauuay triangulation techniques. The singular elements such that the mid-point nodes near crack fornt are shifted at the quarter-points are automatically placed along the 3D crack front. THe complete finite element (FE) model generated, i.e the mesh with material properties and boundary conditions is given to one of the commercial FE codes, and a stress analysis is performed. The SIFs are calculated using the displacement extrapolation method. To demonstrate practical performance of the present system, a semi- elliptical surface crack in a plate subjected to tension is solved.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2003.06a
• /
• pp.1824-1827
• /
• 2003

The sheet metal shearing process is normally used in the precision elements such as semi-conductor components. In precision elements, burrs usually reduce the quality of machined parts and cause interference, jamming and misalignment during assembly procedures and because of their sharpness, they can be safety hazard to personnel. Furthermore, not only burrs are hard to predict and avoid, but also deburring, the process of removing burrs, is time-consuming and costly. In order to get the burr-free parts, therefore, we developed the precise burr measuring system using the laser. The laser burr measuring system consists of the laser probe, the photo detector, the achromatic doublet lens, and the rotary ＆ the X-Y table. In previous reports, we used simple vertical measuring method. But, as we used relatively bigger laser spot diameter and had the limited reflection angle, it was difficult to obtain the precise measuring results. So called, the spot size effect makes the profile of burr measured distorted and the burr height measured smaller. By introducing the novel laser measuring method which employing the achromatic lens system and the tilting mechanism, we could make the spot size smaller and get the appropriate beam direction angle. Through the experiments, the accuracy of the developed system is proved. The burr height measured during the punching process can be used for automatic deburring and in-situ aligning.

• Annual Conference on Human and Language Technology
• /
• 2014.10a
• /
• pp.193-196
• /
• 2014

본 논문에서는 사전자원에 기반한 한국어 워드넷(Open Korean WordNet: KWN)의 반자동 구축 방법을 제안한다. 제안한 방법에서는 각 전문분야별로 분류된 영어-한국어 대역사전, 일본어-한국어 대역사전을 이용하여 영어 워드넷(Princeton WordNet 3.0)과 일본어 워드넷(Japanese WordNet 1.1)의 어휘를 번역하였다. 그리고 번역 결과의 애매성을 해소하기 위하여, (1)영어와 일본어에 대한 한국어 대역어의 중복 여부, (2)사전의 분야 정보와 워드넷의 계층구조를 고려하였다. 제안한 방법으로 117,659 개의 워드넷 synset 중 63,221 개(약 54 %)의 synset에 대한 자동번역을 수행하여 한국어 워드넷을 구축하였다. 그리고 워드넷 synset의 정의문은 한국어 사전의 정의문을 참조하여 한글화 할 수 있도록 하고, 이 과정을 지원하기 위한 정의문 추천 알고리즘을 제안한다. 제안한 방법에 기반하여 전문가들이 상호 협력하여 한국어 워드넷을 구축할 수 있는 시스템을 개발한다.

• The Bulletin of The Korean Astronomical Society
• /
• v.42 no.1
• /
• pp.34.3-35
• /
• 2017

Spectral energy distribution camera for QUasars in EArly uNiverse (SQUEAN) is a successor of Camera for Quasars in EArly uNiverse (CQUEAN) which was developed by Center for the Exploration of the Origin of the Universe and operated at the 2.1 m Otto Struve Telescope in the McDonald Observatory, USA, since 2010. The software of SQUEAN controls a science camera, a guiding camera, and a filter wheel, and communicates with the telescope control system (TCS). It has been constantly revised and modularized according to the upgrades of the TCS and the hardware changes. Recently we have implemented the stable network communication and the semi-automatic focusing modules to enhance observational convenience. In this presentation we describe the current status of the SQUEAN control software and introduce a software architecture which is optimized on efficient astronomical observations.