• 대한공간정보학회지
• /
• 제16권3호
• /
• pp.111-117
• /
• 2008

본 연구의 목적은 다수의 GPS 센서를 이용하여 장대교량의 거동을 실시간으로 모니터링 하여 분석 할 수 있는 시스템을 개발하는데 있다. 다수의 GPS 센서를 장대교량의 주요 지점에 설치하고 변위를 실시간으로 측정하여 전체적인 거동을 분석한다. 본 연구에서 개발한 거동분석시스템은 교량의 전체적, 지점별 변위정보를 수치화, 시각화하고 이벤트에 대해 능동적인 대응이 가능하며 측정데이터의 DB화를 통해서 거동분석에 관계된 통계정보를 산출할 수 있다.

• 통합자연과학논문집
• /
• 제11권4호
• /
• pp.192-196
• /
• 2018

The mobile agent paradigm can be exploited in a variety of ways, ranging from low-level system administration tasks to middle ware to user-level applications. Mobile agents can be useful in building middle-ware services such as active mail systems, distributed collaboration systems, etc. An active mail message is a program that interacts with its recipient using a multimedia interface, and adapts the interaction session based on the recipient's responses. The mobile agent paradigm is well suitable to this type of application, since it can carry a sender-defined session protocol along with the multimedia message. Mobile agent communication is possible via method invocation on virtual references. Agents can make synchronous, one-way, or future-reply type invocations. Multicasting is possible, since agents can be aggregated hierarchically into groups. A simple check-pointing facility has also been implemented. Another proposed solution is to use multi agent computer systems to access, filter, evaluate, and integrate this information. We will present the overall architectural framework, our agent design commitments, and agent architecture to enable the above characteristics. Besides, the each information needed a mobile agent system such as text, graphic, image, audio and video etc, constructed a great capacity multimedia database system. However, they have problems in establishing connections over multiple subnetworks, such as no end-to-end connections, transmission delay due to ATM address resolution, no QoS protocols. We propose a new connection management scheme in the thesis to improve the connection management involved of mobile agent systems.

• 문화기술의 융합
• /
• 제7권4호
• /
• pp.675-681
• /
• 2021

최근 인공지능 스피커 시장이 성장하면서 사용자와 자연스러운 대화가 가능한 음성합성 기술에 대한 수요가 증가하고 있다. 따라서 다양한 음색의 목소리를 생성할 수 있는 다화자 음성합성 시스템이 필요하다. 자연스러운 음성을 합성하기 위해서는 대용량의 고품질 음성 DB로 학습하는 것이 요구된다. 그러나 많은 화자가 발화한 고품질의 대용량 음성 DB를 수집하는 것은 녹음 시간과 비용 측면에서 매우 어려운 일이다. 따라서 각 화자별로는 소량의 학습 데이터이지만 매우 많은 화자의 음성 DB를 사용하여 음성합성 시스템을 학습하고, 이로부터 다화자의 음색과 운율 등을 자연스럽게 표현하는 기술이 필요하다. 본 논문에서는 화자인식 기술에서 사용하는 딥러닝 기반 x-vector 기법을 적용하여 화자 인코더를 구성하고, 화자 인코더를 통해 소량의 데이터로 새로운 화자의 음색을 합성하는 기술을 제안한다. 다화자 음성합성 시스템에서 텍스트 입력에서 멜-스펙트로그램을 합성하는 모듈은 Tacotron2로, 합성음을 생성하는 보코더는 로지스틱 혼합 분포가 적용된 WaveNet으로 구성되어 있다. 학습된 화자 임베딩 신경망에서 추출한 x-vector를 Tacotron2에 입력으로 추가하여 원하는 화자의 음색을 표현한다.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제34권7호
• /
• pp.585-594
• /
• 2007

오류역전파 방법을 이용하는 신경망들은 패턴들의 학습시간이 매우 오래 걸리고 또한 추가학습과 반복학습의 한계를 가지며, 이런 단점을 보완할 수 있는 이진신경망(Binary Neural Network, BNN)이 Aleksander에 의해 제안되었다. 그러나 BNN도 반복학습에 있어서는 단점을 가지고 있으며, 일반화 패턴을 추출하기 어렵다. 본 논문에서는 BNN의 구조를 개선하여 반복학습과 추가학습이 가능할 뿐 아니라, 특징점들까지 추출할 수 있는 다중 판별자를 가지는 삼차원 뉴로 시스템을 제안한다. 제안된 모델은 기존의 BNN을 기반으로 하여 만들어진 이차원 특징을 가지는 Single Layer Network(SLN)에 귀환회로가 추가되어 특징점들을 누적할 수 있는 삼차원 신경망이다. 학습을 통해 누적된 정보는 판별자의 각 신경세포에 임계치를 조정함으로써 일반화 패턴을 추출할 수 있다. 그리고 생성된 일반화 패턴을 인식에 재사용함으로써 반복학습의 효율성을 높였다. 최종 판정 단계에서는 Maximum Response Detector(MRD)를 이용하였다. 본 논문에서 제안한 시스템을 평가하기 위하여 NIST에서 제공하는 숫자 자료를 이용하였으며, 99.3%의 인식률을 얻었다.

• Journal of Korean Medical Science
• /
• 제33권43호
• /
• pp.239.1-239.12
• /
• 2018

Background: We described a novel multi-step retinal fundus image reading system for providing high-quality large data for machine learning algorithms, and assessed the grader variability in the large-scale dataset generated with this system. Methods: A 5-step retinal fundus image reading tool was developed that rates image quality, presence of abnormality, findings with location information, diagnoses, and clinical significance. Each image was evaluated by 3 different graders. Agreements among graders for each decision were evaluated. Results: The 234,242 readings of 79,458 images were collected from 55 licensed ophthalmologists during 6 months. The 34,364 images were graded as abnormal by at-least one rater. Of these, all three raters agreed in 46.6% in abnormality, while 69.9% of the images were rated as abnormal by two or more raters. Agreement rate of at-least two raters on a certain finding was 26.7%-65.2%, and complete agreement rate of all-three raters was 5.7%-43.3%. As for diagnoses, agreement of at-least two raters was 35.6%-65.6%, and complete agreement rate was 11.0%-40.0%. Agreement of findings and diagnoses were higher when restricted to images with prior complete agreement on abnormality. Retinal/glaucoma specialists showed higher agreements on findings and diagnoses of their corresponding subspecialties. Conclusion: This novel reading tool for retinal fundus images generated a large-scale dataset with high level of information, which can be utilized in future development of machine learning-based algorithms for automated identification of abnormal conditions and clinical decision supporting system. These results emphasize the importance of addressing grader variability in algorithm developments.

• 한국멀티미디어학회논문지
• /
• 제17권8호
• /
• pp.937-945
• /
• 2014

Multi-modal biometric system can use another biometric trait in the case of having deficiency at a biometric trait. It also has an advantage of improving the performance of personal identification by using multiple biometric traits, so studies on new biometric traits have continuously been performed. The inner face of finger is a relatively new biometric trait. It has two major features of knuckle lines and wrinkles, which can be used as discriminative features. This paper proposes a finger identification method based on displacement vector to effectively process some variation appeared in contactless hand image. At first, the proposed method produces displacement vectors, which are made by connecting corresponding points acquired by matching each pair of local block. It then recognize finger by measuring the similarity among all the detected displacement vectors. The experimental results using pubic CASIA hand image database show that the proposed method may be effectively applied to personal identification.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2003년도 춘계학술발표논문집 (중)
• /
• pp.1071-1074
• /
• 2003

Now internet is very popular. Many company services various contents in Web. One of the best way for internet community is network game service for internet member. There are some company to service high quality network game for gathering members. Those company make success for getting fee from their member. Global on-line game market is small but stability growing. Thus Microsoft invest on line game. On this paper, one of a java network turn game system called by "Hoola" is designed and implemented. This Application consist of Client/Server and Database module and multi-tier Architecture, because of easily translation to PDA or mobile phone.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2004년도 Proceedings of ISRS 2004
• /
• pp.618-620
• /
• 2004

This paper explores the use of IKONOS imagery of 1 m resolution panchromatic (PAN) band and 4 m resolution multi-spectral (MS) band in the development of agriculture­related data inventories. Three images (May 25, 2001, December 25, 2001, October 23, 2003) were used to obtain temporal distributions in crop cover characteristics such as rice, pear, grape, red pepper, corn, barley, garlic and surface water cover of reservoir with field investigations. The availability and cost problems are expected to solve by KOMPSAT-2 that is scheduled to launch in 2005. The capability of KOMPSAT-2 image for crop and rural water resources management will increase by accumulating temporal data inventories as a database.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2001년도 제13회 한글 및 한국어 정보처리 학술대회
• /
• pp.530-537
• /
• 2001

본 논문은 자연어 인터페이스에 기반한 관계형 데이터베이스 상에서의 질의 응답 시스템에 대해 기술한다. 본 시스템은 다국어, 다중 도메인, 다중 DBMS를 지원하는 시스템으로, 주로 오디오와 비디오 관련 제품들에 대한 정보를 다룬다. Lexico-semantic pattern (LSP) 문법을 관계형 데이터베이스 상에서의 질의 응답 시스템에 최초로 도입하여 기존의 시스템들에 비해 높은 성능을 보이며, linguistic front-end (LFE)와 database back-end (DBE)를 명확히 구분하고 각종 리소스들을 엔진과 분리함으로써 높은 이식성을 가지도록 한다.

• 한국해양공학회:학술대회논문집
• /
• 한국해양공학회 2000년도 추계학술대회 논문집
• /
• pp.33-37
• /
• 2000

Milling fixture is one kind of machining device according to the industrial demands for multi manufacturing products on the growing at alarming rate. In the field of design and making for machine tool working, welding, assembling with fixture for mass production is a specific division. In order to prevent the production defects, occurring, the optimum design of product, fixture putting in the field is very significant manufacturing method. They require analysis of many kinds of important factors, theory and practice of machine tool operating process and its situations, fixture and its structure, machining condition for tool making tool materials, heat treatment of fixture components, know-how and so on. In this study we designed and constructed a milling fixture of mass production and performed tryout under the AUTO CAD, database, I-DEAS and WINDOW environment. Especially part 1 of this study is reveals with the analysis of part drawing, fixture planning, fixture design etc.