• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제3권2호
• /
• pp.93-100
• /
• 2014

카메라의 시선 방향에 따라 다른 영상을 생성하는 3차원 물체를 2차원 영상만으로 인식하는 것은 어려운 일이다. 특히 영상 생성 시 강한 투영변환(perspective transformation) 이 발생할 경우 투영된 물체의 이미지에 대한 국소 특징을 정의하는 SIFT(Scale-Invariant Feature Transform) 알고리즘은 물체 인식에 한계가 있다. 본 논문에서는 3차원 물체를 하나의 특정 축을 중심으로 회전시키면서 얻은 복수의 영상을 학습 데이터로 활용한 물체인식 방법을 제안한다. 이 방법을 이용하여 복수 영상의 특징 점들을 하나의 특징 공간으로 합성하고 영상들 간의 기하학적인 관계를 이용하여 중복된 영역을 제거한 모델을 생성하면 임의의 3차원 회전이 적용된 물체를 인식할 수 있다. 실험에서는 알고리즘의 유용성을 먼저 확인하기 위해 조명조건과 카메라의 위치를 일정하게 유지하였다. 이 방법에 의해 SIFT 알고리즘만으로 인식이 힘들었던 3차원 물체의 다양한 외관(appearance) 인식이 가능하게 되었다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2008년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.107-108
• /
• 2008

본 논문에서는 8-bits의 그레이레벨 영상에서 그레이레벨(gray-level) 값들의 발생 빈도(frequency)에 따른 순위 변환(Ranking-Transformation)을 이용한 효율적인 무손실 압축을 제안한다. 제안한 기법은 서로 인접한 픽셀의 그레이레벨의 쌍에 대한 발생 빈도를 토대로 원 영상의 그레이레벨 값을 이에 대응하는 순위 값으로 변환시킨다. 이때 부가정보가 발생하지 않도록 입력영상의 픽셀들에 대하여 한 픽셀씩 순차적으로 지정한 순위로 재구성한다. 실험결과, 부가 정보 없이 입력 영상을 압축하게 되어 엔트로피 부호화기를 통한 디지털 영상들의 효율적인 압축 성능 향상을 기대할 수 있다.

• 전자공학회논문지S
• /
• 제35S권2호
• /
• pp.146-156
• /
• 1998

Thispapr proposes hybrid navigation parameter estimation using sequential aerial images. The proposed navigation parameter estimation system is composed of two parts: relative position estimation and absolute position estimation. the relative position estimation recursively computes the current velocity and absolute position estimation. The relative position estimation recursively computes the current velocity and position of an aircraft by accumulating navigation parameters extracted from two succesive aerial images. Simple accumulation of parameter values decreases reliability of the extracted parameters as an aircraft goes on navigating. therefore absolute position estimation is required to compensate for position error generated in the relative position step. The absolute position estimation algorithm combining image matching and digital elevation model(DEM) matching is presented. Computer simulation with real aerial image sequences shows the efficiency of the proposed hybrial algorithm.

• 천문학회지
• /
• 제36권spc1호
• /
• pp.21-27
• /
• 2003

In the present study, we have investigated morphology and evolution of small-scale Ha dynamic features on the quiet sun by analyzing video magnetograms and high resolution Ha images simultaneously taken for 5 hours at Big Bear Solar Observatory on April 18, 1997. From comparisons between time sequential longitudinal magnetograms and H$\alpha$ images covering $150" {\times} 150"$, several small-scale H$\alpha$ dynamic features have been observed at a site of magnetic flux cancellation. A close relationship between such features and cancelling magnetic fluxes has been revealed temporarily and spatially. Our results support that material injection by chromospheric magnetic reconnect ion may be essential in supporting numerous small-scale H$\alpha$ dynamical absorption features, being in line with recent observational studies showing that material injection by chromospheric magnetic reconnect ion is essential for the formation of solar filaments.

• 전기학회논문지
• /
• 제59권12호
• /
• pp.2314-2318
• /
• 2010

For human-robot interaction, a robot should recognize the meaning of human behavior. In the case of static behavior such as face expression and sign language, the information contained in a single image is sufficient to deliver the meaning to the robot. In the case of dynamic behavior such as gestures, however, the information of sequential images is required. This paper proposes behavior classification by using fuzzy classifier to deliver the meaning of dynamic behavior to the robot. The proposed method extracts feature points from input images by a skeleton model, generates a vector space from a differential image of the extracted feature points, and uses this information as the learning data for fuzzy classifier. Finally, we show the effectiveness and the feasibility of the proposed method through experiments.

• 만화애니메이션 연구
• /
• 통권6호
• /
• pp.132-151
• /
• 2002

Animation is an art, which experiences through visual and auditory sense. The purpose of this study is to find out whether the theory of film sound can be also applied to animation sound. The study also aims to find out whether the mood and feeling of animation can be influenced by rhythm and fidelity of sound. To begin with, exampled a few animations and analyzed their sound base on the theoretical background of film sound and found out that influence of rhythm and fidelity of sound in animation is much more effective than in film. To find out the relationship between the changes of the mood according to different rhythm, I have applied the various rhythms of sound in a same sequential images. Result proved that the mood of animation was changed according to different rhythms that applied to animation. sound has a great ability of delivering the mood prior to images and different sound can change the overall feeling and mood of animation. Also proved that influence of fidelity of sound is much greater than rhythms of sound, especially in comic animation.

• 한국멀티미디어학회논문지
• /
• 제16권12호
• /
• pp.1439-1445
• /
• 2013

We present a Taekwondo training system using a hybrid sensing technique of a body sensor and a visual sensor. Using a body sensor (accelerometer), rotational and inertial motion data are captured which are important for Taekwondo motion detection and evaluation. A visual sensor (camera) captures and records the sequential images of the performance. Motion chunk is proposed to structuralize Taekwondo motions and design HMM (Hidden Markov Model) for motion recognition. Trainees can evaluates their trial motions numerically by computing the distance to the standard motion performed by a trainer. For motion training video, the real-time video images captured by a camera is overlayed with a visualized body sensor data so that users can see how the rotational and inertial motion data flow.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2006년도 Proceedings of ISRS 2006 PORSEC Volume II
• /
• pp.946-949
• /
• 2006

With repeated coverage, spaceborne SAR (Synthetic Aperture Radar) instruments provide the most efficient means to monitor and study the changes in important elements of the marine environment. Due to highresolution of SAR data, the coverage of SAR sensor is always limited, especially for a repeat cycle. With more SAR sensors from various satellites, new data products such as ocean surface drift can be derived when two SARs' tracks overlap in a short time over coastal areas. Currently, there are two SAR sensors on different satellites with almost the exactly same path. That is, ERS-2 is following ENVISAT with a 30-minutes delay, which will be a good timing for ocean mesosclae feature tracking. For another application, a mystery ship near a big eddy with strong ship wake has been tracked between ERS-2 and ENVISAT SAR images to estimate its ship speed.

• 대한자기공명의과학회:학술대회논문집
• /
• 대한자기공명의과학회 2001년도 제6차 학술대회 초록집
• /
• pp.136-136
• /
• 2001

Purpose： Subclavian steal refers to the retrograde flow of blood in a vertebral artery that supplies t ipsilateral shoulder and arm caused by proximal subclavian artery stenosis or occlusion. T purpose of this exhibit is to demonstrate MR findings of subclavian steal on contrast-enhanc 3D (CE 3DMRA) and 2D TOF MR angiography. Method： Four patients(men 3 and women 1, age： 28-78years) with subclavian steal obtained both CE 3DMRA and digital subtraction angiography(DSA) including subclavian arterie Sequential imaging was undertaken during first pass after double dose of Gd-DTPA (0 mM/kg) Injected by a power injector. Coronal source images were obtained with coronal D-fast low angle shot sequence(TR/TE/flip angle=3.8/l.3/35, acquisition time= 10sec/one measurement). Precontrast imaging was subtracted from enhanced images and maximu intensity projection was done. 2D time-of-flight MR angiography(2D TOF) of the carot bifurcation was added in all cases with post-saturation. All studies were review retrospectively.

• 한국가시화정보학회지
• /
• 제11권3호
• /
• pp.12-16
• /
• 2013

The impact and spreading behaviors of silicon dioxide nanoparticle colloidal suspension droplets were quantitatively visualized using a high-speed imaging system. Millimeter-scale droplets were generated by a syringe pump and a needle. Droplets of different velocity were impacted on a non-porous solid surface. Images were consecutively recorded using a CMOS high-speed camera at 5000 fps (frames per second) for millimeter-scale droplets. Temporal variations of droplet diameter, velocity and maximum spreading diameters were evaluated from the sequential images captured for each experimental condition. Effects of Reynolds number, Weber number, and particle concentration were investigated experimentally.