• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2007.04a
• /
• pp.115-119
• /
• 2007

종중복도 60%이상, 횡중복도 30%이상 촬영되어지는 항공영상과 스테레오 촬영이 가능한 위성영상은 도화에 사용되는 입체시를 이용하여 영상(Stereo Aerial Image) 자체를 화면상에 입체적으로 구현하여 건물의 높이 정보 판독 및 해당 지역상에서 년도별 변화지역을 판독하는 일련의 업무 수행에 있어 중요한 자료로 활용하고 있지만 장기간의 작업시간, 작업에 대한 정확성에 취약점을 나타내고 있으며, 이는 행정업무의 효율성 저하요인을 발생하고 있다. 이에 본 연구에서는 이러한 항공사진 및 위성영성의 촬영상의 특성을 활용하여 영상 매칭 DEM을 활용한 높이정보의 변화와 영상 정합을 통한 변화지역 판독을 자동화 하는 시스템을 구현하였다. 시스템 구현을 위해서 개발 언어로 Visual C++을 사용하였으며, 개발된 알고리즘에 대한 평가 수행을 위해 사용자가 직접 입체 판독 및 분석을 수행할 수 있도록 편광 모니터를 사용하여 판독 시스템을 추가적으로 개발하였다.

• Korean Journal of Construction Engineering and Management
• /
• v.12 no.5
• /
• pp.137-145
• /
• 2011

BIM Data exchange using standard format can provide a user friendly and practical way of integrating the BIM tools in the life cycle of a building on the currently construction industry which is participated various stakeholder. It used IFC format to exchange the BIM data from Design software to energy analysis software. However, since we can not use the material name data in the library of an energy analysis directly, it is necessary to input the material property data for building energy analysis. In this paper, to matching the material named of name of DOE-2 default library, rhe extracted material names from BIM file are inferred by the ontology With this we can make the reliable input data of the engine by development a standard data and also increase the efficient of building energy analysis process. The methodology can enable to provide a direction of BIM-based information management system as a conceptual study of using ontology in the construction industry.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.161-163
• /
• 2021

In an apartment or private building, it is difficult for the user to operate the elevator button in a similar situation with luggage in both hands. In an environment where human contact must be minimized due to a highly infectious virus such as COVID-19, it is inevitable to operate an elevator based on untact. This paper proposes an operating system capable of operating the elevator by using the user's voice and image processing through the user's face without pressing the elevator button. The elevator can be operated to a designated floor without pressing a button by detecting the face of a person entering the elevator by detecting the person's face from the camera installed in the elevator, matching the information registered in advance. When it is difficult to recognize a person's face, it is intended to enhance the convenience of elevator use in an untouched environment by controlling the floor of the elevator using the user's voice through a microphone and automatically recording access information.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.9
• /
• pp.1152-1157
• /
• 2021

This paper proposes a face recognition method for access control of apartment building. Different from most existing face recognition methods, the proposed one does not require any manual process for face enrollment. When a person is exiting through the main entrance door, his/her face data (i.e., face image and face feature) are automatically extracted from the captured video and registered in the database. When the person needs to enter the building again, the face data are extracted and the corresponding face feature is compared with the face features registered in the database. If a matching person exists, the entrance door opens and his/her access is allowed. The face data of the matching person are immediately deleted and the database has the latest face data of outgoing person. Thus, a higher recognition accuracy could be expected. To verify the feasibility of the proposed method, Python based face recognition has been implemented and the cloud service provided by a web portal.