• 인터넷정보학회논문지
• /
• 제21권3호
• /
• pp.113-121
• /
• 2020

Human Pose Estimation (HPE) which localizes the human body joints becomes a high potential for high-level applications in the field of computer vision. The main challenges of HPE in real-time are occlusion, illumination change and diversity of pose appearance. The single RGB image is fed into HPE framework in order to reduce the computation cost by using depth-independent device such as a common camera, webcam, or phone cam. However, HPE based on the single RGB is not able to solve the above challenges due to inherent characteristics of color or texture. On the other hand, depth information which is fed into HPE framework and detects the human body parts in 3D coordinates can be usefully used to solve the above challenges. However, the depth information-based HPE requires the depth-dependent device which has space constraint and is cost consuming. Especially, the result of depth information-based HPE is less reliable due to the requirement of pose initialization and less stabilization of frame tracking. Therefore, this paper proposes a new method of HPE which is robust in estimating self-occlusion. There are many human parts which can be occluded by other body parts. However, this paper focuses only on head self-occlusion. The new method is a combination of the RGB image-based HPE framework and the depth information-based HPE framework. We evaluated the performance of the proposed method by COCO Object Keypoint Similarity library. By taking an advantage of RGB image-based HPE method and depth information-based HPE method, our HPE method based on RGB-D achieved the mAP of 0.903 and mAR of 0.938. It proved that our method outperforms the RGB-based HPE and the depth-based HPE.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제5권4호
• /
• pp.440-448
• /
• 1999

본 논문은 객체지향형 분산처리 환경 하에서 오디오나 비디오 등과 같은 실시간(real-time) 스트림(stream) 데이타를 처리하는 데 필요한 소프트웨어 기반구조를 설계하고 구현한 내용을 기술한다. 본 논문에서 제시한 DAViS(Distributed Object Framework supporting Audio/Video Streaming)는, 오디오/비디오 데이타의 처리와 관련된 여러 소프트웨어 구성요소들을 분산객체로 추상화하고, 그 객체들간의 제어정보 교환경로와 오디오/비디오 데이타 전송경로를 서로 분리하여 처리한다. 분산응용프로그램 작성자는 DAViS에서 제공하는 서비스들을 이용하여, 기존의 분산프로그래밍 환경이 제공하는 것과 동일한 수준에서 오디오/비디오 데이타에 대한 처리를 표현할 수 있다. DAViS는, 새로운 형식의 오디오/비디오 데이타를 처리하는 부분을 손쉽게 통합하고, 하부 네트워크의 전송기술이나 컴퓨터시스템 관련 기술의 진보를 신속하고 자연스럽게 수용할 수 있도록 하는 유연한 구조를 가지고 있다. Abstract This paper describes the design and implementation of software framework which supports the processing of real-time stream data like audio and video in distributed object-oriented computing environment. DAViS(Distributed Object Framework supporting Audio/Video Streaming), proposed in this paper, abstracts software components concerning the processing of audio/video data as distributed objects and separates the transmission path of data between them from that of control information. Based on DAViS, distributed applications can be written in the same abstract level as is provided by the existing distributed environment in handling audio/video data. DAViS has a flexible internal structure enough to easily incorporate new types of audio/video data and to rapidly accommodate the progress of underlying network and computer system technology with very little modifications.

• 한국컴퓨터정보학회논문지
• /
• 제26권11호
• /
• pp.51-57
• /
• 2021

본 논문에서는 립모션 디바이스를 이용하여 사용자의 손 움직임을 계산하고, 이로부터 저글링 동작뿐만 아니라 이것을 활용하여 팔 근육을 연습 및 분석할 수 있는 새로운 프레임워크를 제안한다. 제안하는 방법은 실시간으로 사용자의 손동작에 맞춰 가상환경에서 공의 움직임을 매핑 할 수 있고, 근육의 이완과 수축을 시각화하여 운동량을 분석할 수 있다. 제안하는 프레임워크는 크게 세 부분으로 구성된다 : 1) 립모션 디바이스를 이용하여 사용자의 손 위치를 추적한다. 2) 저글링을 하듯이, 사용자가 공을 던지는 행동 패턴을 이벤트로 정의한다. 3) 사용자의 손 위치를 기준으로 저글링 형태의 움직임을 공에 매핑하기 위한 포물선 기반 입자 기법을 제안한다. 결과적으로 본 논문의 프레임워크를 이용하면 실시간 저글링 게임을 할 수 있다.

• 한국시뮬레이션학회논문지
• /
• 제19권3호
• /
• pp.45-54
• /
• 2010

반도체 제조공정에 내재된 복잡성은 작업일정(job scheduling) 문제를 해석적 방법으로는 풀기 어렵기 때문에 보통 시스템 파라미터의 변화에 대한 효과를 이산사건 모델링 시뮬레이션에 의존하여 왔다. 한편 장비 고장 등 예측 불가능한 사건들은 고정된 작업일정 기법을 사용할 경우 전체 공정의 효율을 악화시킨다. 따라서 이러한 불확실성에 대해 최적의 성능을 내기 위해서는 작업일정을 실시간으로 대처 변경하는 것이 필요하다. 본 논문은 반도체 제조 공정에 대해 시스템 제어관점의 접근방법을 적용하여 이 문제에 적응형 실시간 작업제어 틀을 제안하고, DEVS 모델링 시뮬레이션 환경을 기반으로 제안된 틀을 설계 구현하였다. 제안된 방법은 기존의 임기응변적인 소프트웨어적인 방법에 비추어볼 때 전체 시스템을 이해하기 쉬우면서도 또한 추가되는 작업제어 규칙도 쉽게 추가 적용할 수 있는 유연성을 장점으로 가지고 있다. 여러 가지 실험결과 제안된 적응형 실시간 작업제어 프레임워크는 고정 작업규칙 방법에 비해 훨씬 나은 결과를 보여주어 그 효용성을 입증하였다.

• International Journal of Control, Automation, and Systems
• /
• 제6권1호
• /
• pp.62-75
• /
• 2008

Vision-based simultaneous localization and map building using a single camera, while compelling in theory, have not until recently been considered extensive in the practical realm of the real world. In this paper, we propose a simple framework for the monocular SLAM of an indoor mobile robot using natural line features. Our focus in this paper is on presenting a novel approach for modeling the landmark before integration in monocular SLAM. We also discuss data association improvement in a particle filter approach by using the feature management scheme. In addition, we take constraints between features in the environment into account for reducing estimated errors and thereby improve performance. Our experimental results demonstrate the feasibility of the proposed SLAM algorithm in real-time.

• 대한치과보철학회지
• /
• 제34권4호
• /
• pp.869-902
• /
• 1996

Flexion of a metal/ceramic fixed partial denture(EPD) frameworks under function can cause fracture of porcelain or deterioration of the cement seal. This study evaluated the flexion characteristics of three-unit mandibular FPD frameworks, repacing the second pre-molar under compressive load(200g, 400g). Testing was accompished with real-time holographic interferometry, using 6 porcelain fused-to metal frameworks. Tested alloys were non-precious alloy(Heracles, Holland), semi-precious alloy(Degudent U, Germany) and precious alloy(Degudent H, Germany). Changes of the fringe patterns according to the heat treatment(porcelain firing cycle), various loads(200g, 400g), occlusal forms(occlusal porcelain veneering, facial porcelain veneering), various alloys and post-soldering units were compared. Dental study model(Nissan dental products, Inc. D51DP-500A, Japan) and six 3-unit metal/ceramic fixed partial denture frameworks were used as experimental materials. 36 holograms were taken on fixed dental study model by using the 10mW He-Ne laser and real-time holographic interferometry. On the basis of this study, the following conclusions can be drawn : 1. In the frameworks for facial porcelain veneering, the semi-precious alloy framework was least deformed and precious alloy framework, non-precious alloy framework orderly before heat treatment, and the deformation was not shown great difference among three alloys after heat treatment and post-soldering. 2. In the frameworks for occlusal porcelain veneering, the precious alloy framework was greatest deformed and the deformation was not difference between semi-precious alloy framework and non-precious alloy framework before, after heat treatment, and the deformation was not shown great difference among three alloys after post soldering. 3. In the non-precious alloy frameworks for facial porcelain veneering and occlusal porcelain veneering, the deformation was greatly decreased after heat treatment and conversely increased after post-soldering. 4. In the semi-precious alloy framework for facial porcelain veneering, the deformation was not detectable after heat treatment and increased after post-solder. And in the frame-work for occlusal porcelain veneering, the deformation was slightly decreased after heat treatment and increased after post-soldering. 5. In the precious alloy framework for facial porcelain veneering, the deformation was greatly decreased after heat treatment and increased after post-soldering, And in the framework for occlusal porcelain veneering, the deformation was greatly decreased after heat treatment and decreased after post-soldering.

• 정보보호학회논문지
• /
• 제24권4호
• /
• pp.669-679
• /
• 2014

본 연구에서는 스마트금융 서비스를 위한 서버 측 보안 프레임워크를 제안한다. 국내 금융기관의 전자금융서비스를 제공하는 서버 측 프레임워크 구조를 살펴보면 대부분 서비스제공위주의 구조를 가지고 있다. 따라서, 보안관련 요구사항들은 비즈니스 로직들에 같이 포함되어 있는 경우가 대부분이기 때문에 보안 사고에 효과적으로 대응하기 어렵다. 본 논문에서는 전자금융서비스시 보안영역을 비즈니스영역과 분리하여 업무에 대한 의존도(Dependency) 없이 보안 정책을 실시간으로 적용할 수 있는 프레임워크를 제안한다. 이를 통하여 보안관련 위협에 대한 신속하고 효과적인 대응기반을 제시한다. 또한 현재 서비스하고 있는 시스템구조에서도 시스템의 큰 변경없이 제안 프레임워크를 적용할 수 있는 방안을 제시한다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제7권4호
• /
• pp.221-227
• /
• 2007

This paper introduces an on-line particle-filtering-based framework for failure prognosis in nonlinear, non-Gaussian systems. This framework uses a nonlinear state-space model of the plant(with unknown time-varying parameters) and a particle filtering(PF) algorithm to estimate the probability density function(pdf) of the state in real-time. The state pdf estimate is then used to predict the evolution in time of the fault indicator, obtaining as a result the pdf of the remaining useful life(RUL) for the faulty subsystem. This approach provides information about the precision and accuracy of long-term predictions, RUL expectations, and 95% confidence intervals for the condition under study. Data from a seeded fault test for a UH-60 planetary carrier plate are used to validate the proposed methodology.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2008년도 International Symposium on Remote Sensing
• /
• pp.155-158
• /
• 2008

The needs for digital models of real environment such as 3D terrain or cyber city model are increasing. Most of applications related with modeling and simulation require virtual environment constructed from geospatial information of real world in order to guarantee reliability and accuracy of the simulation. The most fundamental data for building virtual environment, terrain elevation and orthogonal imagery is acquired from optical sensor of satellite or airplane. Providing interoperable and reusable digital model is important to promote practical application of high-resolution satellite imagery. This paper presents the new research regarding representation of geospatial information, especially for 3D shape and appearance of virtual terrain, and describe framework for constructing real-time 3D model of large terrain based on high-resolution satellite imagery. It provides infrastructure of 3D simulation with geographical context. Details of standard-based approach for providing infrastructure of real-time 3D simulation using high-resolution satellite imagery are also presented. This work would facilitate interchange and interoperability across diverse systems and be usable by governments, industry scientists and general public.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권10호
• /
• pp.4823-4843
• /
• 2017

In this paper, an improved utility-based cross-layer dynamic subcarrier allocation (DSA) algorithm is proposed for single carrier frequency division multiple access (SC-FDMA) system, which adopts adaptive service rate control (ASRC) to eliminate the service rate waste and improve the spectral efficiency in heterogeneous network including non-real-time traffic and real-time traffic. In this algorithm, furthermore, a first in first out (FIFO) queuing model with finite space is established on the cross-layer scheduling framework. Simulation results indicate that by taking the service rate constraint as the necessary condition for optimality, the ASRC algorithm can effectively eliminate the service rate waste without compromising the scheduling performance. Moreover, the ASRC algorithm is able to further improve the quality of service (QoS) performance and transmission throughput by contributing an attractive performance trade-off between real-time and non-real-time applications.