• Proceedings of the Korea Contents Association Conference
• /
• 2017.05a
• /
• pp.13-14
• /
• 2017

딥러닝을 비롯한 전반적인 인공지능에 대한 관심이 뜨겁다. 특정 영역에 영향력을 주었던 과거와 다르게 인공지능의 영향력은 인류문명 전체에 변화를 주고 있다. 예술 분야도 영향을 받고 있는데, 그 중 한 분야는 과학적 실험의 자료를 어떻게 시각화 하느냐의 문제를 풀다가 나오기도 한다. 자료를 시각화하는 것은 실험과정 및 결과를 과학자 및 독자들에게 쉽게 전달하기위한 것이다. 그런데, 그 시각화된 영상 중에는 미적인 아름다움이 있는 경우가 있다. 본 연구자는 강화학습의 정책이 어떻게 개선되고 있는지 보기위해 강화학습의 과정을 시각화 해서 검증하는 시도를 했다. 그 과정에서 만든 자료가 미술적인 관점에서도 아름다움이 있는 작품을 만들 수 있다는 확신이 들어서 강화학습용 디지탈예술 도구를 만들어 작품을 생성해 보았다.

• Journal of Korea Multimedia Society
• /
• v.7 no.5
• /
• pp.617-632
• /
• 2004

Video Data includes moving objects which change spatial positions as time goes by. In this paper, we propose a new modeling method for a moving object contained in the video data. In order to effectively retrieve moving objects, the proposed modeling method represents the spatial position and the size of a moving object. It also represents the visual features and the trajectory by considering direction, distance and speed or moving objects as time goes by. Therefore, It allows various types of retrieval such as visual feature based similarity retrieval, distance based similarity retrieval and trajectory based similarity retrieval and their mixed type of weighted retrieval.

• Journal of KIISE:Databases
• /
• v.31 no.3
• /
• pp.219-231
• /
• 2004

Recently, together with increasing use of multimedia data, many works on moving objects in video databases have been made. Moving objects change visual features and spatial positions with the lapse of time in video data. And they arc related to the other objects or events. In this paper, we propose a new modeling and various query types of moving objects for content based retrieval in video databases. The proposed modeling represents visual features, moving trajectories and semantic contents related to objects. Therefore, it allows to process various query types. And we propose various query operators for the retrieval types. To show the superiority of our modoling, we implement the retrieval systems and compare it with the existing methods in terms of the supporting query types. The proposed method supports various query types and improves the efficiency of the query processing over the existing methods.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.01a
• /
• pp.149-151
• /
• 2014

주의(attention)를 어떻게 할당하는가는 정신물리학뿐만 아니라, 컴퓨터 시각과정을 모델링하는데 중요한 주제 중 하나이다. 기존 연구는 saliency와 같은 활성화 값에 의해서 주의탐사 순위가 결정된다. 본 논문에서는 주의탐사과정을 병렬처리를 통한 활성화 값 추출과정과 순차적 처리를 통한 공간적 전략과정으로 구분하였다. 단서패러다임에 기초한 계산모형을 이용하여, 실제 인간의 수행결과를 AUC와 Levenshtein 척도를 이용하여 비교하였다. Fixation point 비교에서는 인간과 활성화를 기반으로 한 계산모형의 수행은 높은 상관성을 가지고 있었다. 주의궤적 혹은 scanpath 분석에서는 활성화기반 전략보다는 공간적 전략 모형이 더 높은 유사성을 보였다. 이는 주의탐사과정이 병렬처리과정을 통해 얻어진 saliency에 의해서만 결정되는 것이 아니라, 목표물간의 근접성 등의 공간적 전략을 통해 순차적 (sequential) 경로가 생성됨을 의미한다.

• Journal of the Korea Society of Computer and Information
• /
• v.9 no.4 s.32
• /
• pp.85-91
• /
• 2004

The use of gestures provides an attractive alternate to cumbersome interface devices for human-computer interaction. This has motivated a very active research area concerned with computer vision-based analysis and interpretation of hand gestures The most important issues in gesture recognition are the simplification of algorithm and the reduction of processing time. The mathematical morphology based on geometrical set theory is best used to perform the processing. The key idea of proposed algorithm is to track a trajectory of center points in primitive elements extracted by morphological shape decomposition. The trajectory of morphological center points includes the information on shape orientation. Based on this characteristic we proposed the morphological gesture sequence recognition algorithm using feature vectors calculated to the trajectory of morphological center points. Through the experiment, we demonstrated the efficiency of proposed algorithm.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.11
• /
• pp.195-205
• /
• 2013

The analysis of electroencephalographic signal associated with the attention is essential for the understanding of human cognition. In this paper, the characteristic differences between the attention and inattention status in the brain were inspected by nonlinear analysis. The recurrence quantification analysis was applied to the relatively small number of samples of evoked potential having time varying characteristics, where the recurrence plot (RP), the color recurrence plot (CRP), and mean and time-sequential trend parameters were extracted. The dimension and the time delay in phase transformation can be determined by the paired set of extracted parameters. It is observed from RP, CRP, and parameters that the brain dynamics in attention is more complex than that in the inattention, as well as the synchronized brain response is stable in the mean sense but locally time varying. It is feasible that the non-linear analysis method can be useful for the analysis of complex brain dynamics associated during visual attentional task.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1993.10a
• /
• pp.588-592
• /
• 1993

본 연구에서는 로봇의 교시, 궤적계획, 충돌, 로봇동작, 제어알고리즘 등의 평가에 유용하게 사용할 수 있는 다기능 오프라인 프로그래밍 시스템인 POLPS(Pusan_national_university's Off-Line Programming System) 를 개발하였다. POLPS은 4축 스카라형 FARA 로봇을 대상으로 개발되었고, 편리한 방법으로 교시작업을 수행할 수 있으며, 교시 내용에 따라 궤적계획을 수행한 후 로봇을 동작시켜 충돌을 회피하는 동적 시뮬레이션을 수행할 수 있다. 제안된 여러 제어알고리즘 중에서 주어진 작업에 적합한 제어알고리즘을 효율적으로 선정하기 위한 성능의 비교와 평가 기능을 갖추고 있는 것이 주요한 특징이다. 또한 시각위치변경, 온선. 온면제거 및 음영처리, 고속 애니메이션, 환경구성, 인터프리터 명령어 등의 다양한 기능으로 편리하게 사용할 수 있다.

• Journal of Conservation Science
• /
• v.37 no.2
• /
• pp.90-100
• /
• 2021

Outdoor cultural buildings and their accessories receive different amounts of solar radiation depending on their location's latitude, azimuth, and tilt. Shading is also affected by the surrounding terrain and objects, necessitating individual and quantitative shading analysis. In July 2019, this study conducted a shading analysis on the tops, midpoints, and bottoms of wooden pillars in the azimuth of Cheongpunggak, a traditional building in South Korea's National Research Institute of Cultural Heritage. The shading analysis found that the solar access/shade predicted by the solar trajectory meter was 30 minutes slower than measured in the field. The highest solar access and solar radiation levels came from the south, followed by the west, east, and north. The pillars' bases received the highest solar access and solar radiation, followed by their midpoints and tops. Solar access was high at tilt 90°, but solar radiation was high at tilt 0°, due to the light-collection efficiency and the irradiance. Shading on the pillars' tops was caused by the roof eaves, while shading on the midpoints and bases were affected by the surrounding pillars, topography, and other objects. Simultaneous solar access at the tops, midpoints, and bottoms was possible for 365 days for the northwest, west, and southwest pillars but only from October to March for the south and southeast pillars.

• Journal of KIISE:Computing Practices and Letters
• /
• v.6 no.5
• /
• pp.536-542
• /
• 2000

This paper proposes Visual Script Language which can direct a type of motion to 3D human model and create by dragging gesture like as we can express a certain meaning with hand gestures. Traditional motion control technique of articulated figures such as human needs a complex task that draws on highly developed human skills. So it will reduce the amount of motion specification to provide the motion control method that allow users to describe characters' motion at the higher level abstraction. Visual script is the visual gestures to direct various human motions, so users can express the spatial attributes of a motion such as the path of moving with high-level concepts if they use visual script. And we can show that it is possible to control the motion of human model directly and intuitively by development of 3D human motion control system based on visual script.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.46 no.1
• /
• pp.68-77
• /
• 2018

In order to analyze the overall characteristics of the lunar orbiter, the Variable Coast method, which can be launched everyday, is applied to the 3.5 phasing loop transfer trajectory. The mission scenario for the entire process from launching to entering the lunar orbit is set up and performed simulation by selecting the launch pad and launch vehicle. In particular, the SEM(Satellite-Earth-Moon) angle defined in Earth-Moon rotating frame is an important constraint to comprehensively evaluate the 3.5 phasing loop transfer trajectory. The simulation using SEM angle is analyzed from various viewpoints such as launch epoch, coast duration, perigee altitude and ${\Delta}V$ not only trans-lunar trajectory but lunar orbit insertions and the optimum SEM angle is suggested in this study. It is expected that this results will be helpful to evaluate the characteristics of the 3.5 phasing loop transfer trajectory according to the launch vehicle selection by comparison with Fixed Coast analysis results in the future.