• Journal of Korea Society of Industrial Information Systems
• /
• v.26 no.4
• /
• pp.11-18
• /
• 2021

We propose a method that combines learning in a virtual environment and a real environment for indoor autonomous driving through reinforcement learning. In case of learning only in the real environment, it takes about 80 hours, but in case of learning in both the real and virtual environments, it takes 40 hours. There is an advantage in that it is possible to obtain optimized parameters through various experiments through fast learning while learning in a virtual environment and a real environment in parallel. After configuring a virtual environment using indoor hallway images, prior learning was carried out on the desktop, and learning in the real environment was conducted by connecting various sensors based on Jetson Xavier. In addition, in order to solve the accuracy problem according to the repeated texture of the indoor corridor environment, it was possible to determine the corridor wall object and increase the accuracy by learning the feature point detection that emphasizes the lower line of the corridor wall. As the learning progresses, the experimental vehicle drives based on the center of the corridor in an indoor corridor environment and moves through an average of 70 steering commands.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.785-789
• /
• 2008

증강현실은 실제환경에 가상물체를 합성하는 기술을 이용하고 여러가지 상황정보를 반영하여 가상물체가 실제환경에 있는듯한 느낌을 준다. 기존의 연구에서는 주로 가상물체의 좌표계를 실제환경과 일치시킴으로서 사실성을 주고 있다. 하지만 실제 환경의 주변광원이 가상물체에 사실적으로 적용되지 않으면 몰입감이 떨어지게 된다. 본 논문에서는 증강현실에서 가상물체에 실제환경의 상황정보 중 음영정보를 손쉽게 일치시키는 방법을 제안하여 사실성을 증가 시키고자 하였다. 음영효과는 주변광에 의해서 만들어지므로 가상물체를 합성할 위치 주변의 광원을 미러볼을 사용하여 쉽고 간단하게 획득하고 가장물체에 적용될 광원지도를 생성하는 방법을 제안한다. 또한, 본 논문에 소개될 주변광원 정보 획득 및 음영을 적용 방법은 특수장비가 필요없고 실시간으로 처리가 가능하기 때문에 유비쿼터스 환경, 모바일 환경 등 폭 넓게 적용될 수 있다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.8
• /
• pp.477-485
• /
• 2006

In this paper, video avatar, made from live video streams captured from a real participant, was used to represent a virtual participant. By using video avatar to represent participants, the sense of reality for participants can be increased, but the correct registration is also an important issue. We configured the real and virtual cameras to have the same characteristics in order to register the video avatar. Comparing real and synthetic images, which is possible because of the similarities between real and virtual cameras, resolved registration between video avatar captured from real environment and virtual environment. The degree of incorrect registration was represented as energy, and the energy was then minimized to produce seamless registration. Experimental results show the proposed method can be used effectively for registration of video avatar.

• Proceedings of the KAIS Fall Conference
• /
• 2007.11a
• /
• pp.256-258
• /
• 2007

내장 소프트웨어는 실제 시스템의 자원과 원격지의 시스템의 환경에 민감하므로 실제 시스템과 동일한 환경에서 디버깅해야 한다. 그러나 대부분의 내장 소프트웨어를 탑재하는 실제 시스템은 시스템 상태를 조사하거나 제어하는 것이 제한되어 있는 소프트웨어를 디버깅하는 것은 매우 어렵다. 본 논문에서는 원격지의 USB와 실제 시스템의 JTAG을 기반으로 내장 소프트웨어를 디버깅할 수 있는 어뎁터 제안한다. 본 논문은 실제 시스템내의 내장 소프트웨어를 디버깅할 수 있는 경제적인 인터페이싱 환경을 제공한다.

• Environmental engineer
• /
• v.18 s.192
• /
• pp.54-58
• /
• 2002

• Environmental engineer
• /
• v.18 s.193
• /
• pp.54-58
• /
• 2002

• Journal of The Korean Association For Science Education
• /
• v.32 no.10
• /
• pp.1567-1579
• /
• 2012

This study carried a survey to investigate teacher's cognition on preference and actuality about science laboratory and class environment targeting 262 science teachers. The results of this study are as follow: First, the actuality cognition of science teachers on science laboratory and class environment was lower than preference (p<.05). Second, there were no differences between preference and actuality regardless of gender (p<.05). However, the cognition on all of subordinates of preference appeared higher in females than males (p<.05). Third, at all levels of schools, preference is higher than actuality for science laboratory and teaching environment (p<.05). In case of preference, all of the subordinates indicated the difference between elementary and high school teachers (p<.05). On the other hand, in actuality there was a difference between elementary and middle school teachers in 'science laboratory facilities condition' domain only (p<.05). Fourth, the preference was higher than actuality in all school locations (p<.05). And in case of preference, there was no difference in all subordinates regardless of school sites. Whereas the cognition of small-medium city teachers was lower than metropolitan in actuality on the three domains of 'science laboratory facilities condition,' 'teaching condition and service support,' and 'staff policy and practice' (p<.05). As a result, this study informs that upgrading is necessary to achieve inquiry activity in science class in overall teaching environment including science classroom and laboratory.

• Journal of KIISE:Software and Applications
• /
• v.34 no.5
• /
• pp.455-464
• /
• 2007

The construction of a fingerprint database is important to evaluate the performance of an automatic fingerprint recognition system. Due to the cost of collecting fingerprints, there are only few benchmark databases available. Since it is hard to evaluate how robust the system is in various environments with the databases, this paper proposes a novel method that generates fingerprint images automatically from only a few training samples by using the genetic algorithm. Fingerprints generated by the proposed method include similar characteristics of those collected from the corresponding real environment. The proposed method has been verified by comparing with real fingerprint images, showing the usefulness of the method.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.591-593
• /
• 1998

실제감을 가지는 가상의 환경을 만들고, 사용자의 interaction에 맞추어 가상 환경과오브젝트를 조정하는 분야는 최근에 들어 많은 관심을 끌고 있다. 본 논문에서는 실제 영상을 이용하여 3차원 오브젝트 모델링 및 합성하는 방법에 대한 논의를 하였다. 실제 영상에서 깊이 정보를 얻어내는 방법으로는 다해상도 다기선 스테레오 정합 기법(multi-resolution multiple-baseline stereo matching technique : MR-MBS)을 이용하였으며, 실제 영상에서 생성된 모델을 Open Inventor 환경에서 모델을 재구성하고 performer를 이용하여, 최종적인 합성 영상을 만들어 내었다. 합성된 임의시점에서의 영상에서 사용자 조작에 따라 각각의 오브젝트를 조정 할 때 결과 영상을 실시간으로 얻을 수 있었고, 실제 영상의 텍스춰를 그대로 사용함으로써 실제감을 높일 수 있었다.

• Journal of the Korea Society for Simulation
• /
• v.22 no.3
• /
• pp.7-14
• /
• 2013

In this paper, there is a research for providing environmental data to reflect environmental effects to the simulation for the aviation weapon systems by the construction of the synthetic battlefield. The results of the aviation engagement simulation are able to differ by environmental effect. This paper analyzes the real aviation battlefield and designs the synthetic battlefield based on the analysis. In order to construct the designed synthetic battlefield, we collects the real environmental data for the atmosphere and structures the collected data using GIS (Geographic information system interpolation). The main objective of this paper is to design the synthetic battlefield based on the derived environmental factors from the analysis of the real aviation battlefield, and it constructs the designed synthetic battlefield by the collection of real atmosphere data. The constructed synthetic battlefield provides the environmental data which are requested from the distributed simulation system, and it makes the system reflect environmental effects to the simulation.