• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.27 no.1
• /
• pp.735-742
• /
• 2009

National Geographic Institute recently has produced a national paper map by means of a computer aided editing system using a national digital map 2.0. However, the map generalization should be made due to the portrayal difference between the digital map and the paper one and the criterion of the map generalization should be determined by the visual image. The tolerance limit of the map generalization has to be decided based on human vision. For this purpose, this study attempts to measure the size of the building on various scale map and then analyze its result. As a consequence, this study shows us that the building size eligible for human vision should be over 0.4mm in the short side of building on the map. The tolerance limit of an isolated building, a reduced building and a densely built-up area for the map generalization is based on the criterion mentioned above.

• Korea Multimedia Society
• /
• v.10 no.3
• /
• pp.53-63
• /
• 2006

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.10a
• /
• pp.465-469
• /
• 2017

In this paper we propose a human body tracking method for application of smartphones. The conventional human body tracking method is divided into a sensor-based method and a vision-based method. The sensor-based methods have a weakness in that tracking accuracy is low due to cumulative error of position information. The vision-based method has no cumulative error, but it requires reduction of the computational complexity for application of smartphone. In this paper we use the improved HOG algorithm as a human body tracking method for application of smartphone. The improved HOG algorithm is implemented through downsampling and frame sampling. Gaussian pyramid is applied for downsampling, and uniform sampling is applied for frame sampling. We measured the proposed algorithm on two devices, four resolutions, and four frame sampling intervals. We derive the best detection rate among downsampling and frame sampling parameters that can be applied in realtime.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.980-982
• /
• 2022

본 연구는 와상환자에게 자주 발생하는 낙상, 욕창, 불면증을 영상기술을 통해 인식하여 섬망의 조기진단 모형을 구현한다. 실시간 모니터링을 통해 섬망 잠재환자를 선별하고 집중적인 관리와 치료로 이어질 수 있도록 간호인력을 보조하는 데 주된 목적을 두고 있다. 과활동형 섬망은 파생위험 중 하나인 낙상과, 저활동형 섬망은 원인 요소인 욕창과 묶어 자세인식을 통해 판정한다. 또한 주로 밤에 악화되는 섬망의 특성을 고려해 눈 깜빡임을 통한 불면증 검사를 추가로 반영하였다. 낙상과 욕창을 섬망과 묶어 융복합적인 위험예측 시스템을 구축함과 동시에, 기존의 섬망 사정도구들이 지니는 시공간적 제약을 개선함으로써 간호인력의 부담을 덜어줄 것으로 기대된다.

• Proceedings of the IEEK Conference
• /
• 2009.05a
• /
• pp.369-371
• /
• 2009

휴머노이드 로봇에서 스스로 환경을 인식하는 기술은 필수적으로 필요하다. 그러나 정확하게 환경을 인식하기 위해서는 휴머노이드 움직임을 추정하여 보정해야 한다. 따라서 본 논문에서는 type-2 퍼지셋을 이용하여 휴머노이드 로봇의 움직임을 스테레오 비전기반으로 추정하는 방법을 제안하고자 한다. 본 연구에서는 우선 스테레오 비전으로 얻은 Disparity Map을 Type-2 퍼지셋을 이용하여 추적할 대상을 추출한다. 추출된 대상은 보다 정확한 계산을 위하여 Wavelet Transform을 이용하여 정보량을 확장하였으며, 그 결과로 얻어진 결과영상들은 다시 Least suqare approximation과 Type-2 퍼지셋을 이용하여 하나의 결과값으로 나타내어진다. 연속된 두 이미지의 움직임을 추정함으로써 로봇의 움직임을 추정하게 된다. 제안된 방법은 실험을 통하여 그 추정 정확도나 연산속도 면에서 효율적임을 알 수 있었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10b
• /
• pp.421-425
• /
• 2006

최근 들어서 휴머노이드 로봇을 비롯한 로봇에 대하여 관심이 증대되고 있다. 이에 따라, 외모를 닮은 로봇 뿐 만 아니라, 사람과 상호 작용을 할 수 있는 로봇 기술의 중요성이 부각되고 있다. 이러한 상호 작용을 위한 효율적이고, 가장 자연스러운 방법 중의 하나가 비전을 기반으로 한 제스처 인식이다. 제스처를 인식하는데 있어서 가장 중요한 것은 손의 모양과 움직임을 인식하는3차원 제스처 인식이다. 본 논문에서는 3차원 손 제스처를 인식하기 위하여3차원 손 모델 추정 방법과 명령형 제스처 인식 시스템을 소개하고, 수화, 지화 등으로의 확장성을 위한 프레임워크를 제안한다.

• Journal of English Language & Literature
• /
• v.58 no.5
• /
• pp.943-974
• /
• 2012

This paper aims to suggest the compromising vision of nature and technology as the solution to get out of the globally accelerated technology environment in Don DeLillo's Cosmopolis. This novel intends to emphasize on the importance of physical environment as a precondition for the survival of human. Eric wants to be a posthuman with the cybernetic idea, pursuing to be the digital self in a vast biosphere that integrates both the nature and the technology. His obsessive worship of technology through his quest for the futurity results in the effacement of the humanity and the insulation from the nature. Cosmopolis is DeLillo's first 9/11 novel, which describes a young-billionaire asset manager Eric's one-day life in New York in April 2000. Eric can be the third Twin Tower as a symbol of global economic hegemony. By the allusion of the 9/11 catastrophic event, it can be said that Eric's fall is caused by his hubris and avarice as a global capitalist. Crossing the 47th Street toward the West in his limousine, his journey is revealed as the environmental reflections on his desires to attain the futurity and transcendence by technology. This novel cautions that the abuse of technology can bring out the obsolescence and erasure of the humanity and the nature. DeLillo suggests that the best hope for the evolutionary possibility of posthuman can be realized through the correlation with nature and technology. This future-oriented novel warns that the excessive technology should not lead to the disappearance of community and humanity, and the separation of self and nature. It admonishes that they should not follow pseudo-cosmopolitanism as the greedy world citizens, devoting on the velocity of newest technology. This novel recommends that humans should be the world citizen of global ecosystem, making the ameliorative environment through the correlation with self/environment and technology/nature, and gardening the restorative biosphere and the younger planet.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.895-897
• /
• 2005

인간 행동 분석은 컴퓨터 비전 및 패턴인식 분야에서 활발하게 연구가 이루어지는 분야이다. 이러한 행동 분석하고 평가하기 위해서는 다양한 환경과 종류의 제스쳐를 포함하고 있는 데이터베이스의 구축이 필수적이다. 본 논문에서는 총 40명의 사람에 대하여 일상생활에서 일어날 수 있는 14개의 정상 제스처, 위급한 상황에서 발생할 수 있는 10개의 비정상 제스처 그리고 30개의 명령형 제스처를 수집한 KU(Korea University) 제스처 데이터베이스를 소개한다. 각각의 제스처는 스테레오 카메라를 통해 얻어진 2차원 제스처 동영상, 3차원 동작 카메라를 통해 얻어진 3차원 모델의 좌표 정보 그리고 2차원 실루엣 동영상을 포함하고 있다.

• The Journal of the Korea Contents Association
• /
• v.20 no.10
• /
• pp.738-749
• /
• 2020

The stamp commemorating the inauguration of a president is the smallest historical symbol that contains a nation's vision and wishes of the times. The stamp itself is just a collection of images but contains the meaning of the highest leader's perceived image at the time through their face and a symbol. This paper regarded presidential inauguration commemorative stamps in the context of advertisement or promotional materials to structurally analyze the images of the political leaders embedded in stamps utilizing 'semiology' theory, which is considered to be a useful method of advertisement. In accordance with social and cultural changes, the 2000s, when the changes in the presidential inauguration stamp was most distinctively noticeable, was set as the time frame. Only the presidential inauguration commemorative stamps of Moo-hyun Roh, Myung-bak Lee, Geun-hye Park, and Jae-in Moon were selected for this study. In order to interpret the image, a semiotic structure was applied to reconstruct the meaning of the image of the political leaders. Since this can lead to subjective judgment in image analysis, it attempted to find the truth through comprehensive interpretation and approach it from the social and cultural contexts, not a static approach. In this paper, the images of the Korean political leaders in the 2000s shown on stamps are no longer influenced by traditional political factors such as political parties, political ideologies, and regional origins due to the evolution of democracy. This study derives the characteristics of the images of independent and characteristic leaders centered on a person according to the situations and social phenomena of the times.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.3C
• /
• pp.125-132
• /
• 2005

This paper describes a computer vision system based on active IR illumination for real-time gaze discrimination system. Unlike most of the existing gaze discrimination techniques, which often require assuming a static head to work well and require a cumbersome calibration process for each person, our gaze discrimination system can perform robust and accurate gaze estimation without calibration and under rather significant head movement. This is made possible by a new gaze calibration procedure that identifies the mapping from pupil parameters to screen coordinates using generalized regression neural networks (GRNNs). With GRNNs, the mapping does not have to be an analytical function and head movement is explicitly accounted for by the gaze mapping function. Futhermore, the mapping function can generalize to other individuals not used in the training. To further improve the gaze estimation accuracy, we employ a reclassification scheme that deals with the classes that tend to be misclassified. This leads to a 10% improvement in classification error. The angular gaze accuracy is about 5°horizontally and 8°vertically. The effectiveness of our gaze tracker is demonstrated by experiments that involve gaze-contingent interactive graphic display.