• Journal of Korea Multimedia Society
• /
• v.2 no.3
• /
• pp.252-264
• /
• 1999

The key characteristic of video data is its spatial/temporal relationships. In this paper, we propose an content based video retrieval system based on hierarchical data structure for specifying the temporal semantics of video data. In this system, video data's hierarchical structure temporal relationship, inter video object temporal relationship, and moving video object temporal relationship can be represented. We also implemented these video data's temporal relationship into an object-relational database management system using inheritance, encapsulation function overloading, etc. So more extended and richer temporal functions can be used to support a broad range of temporal queries.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.27-29
• /
• 2021

인공지능 기반 머신 비전 응용이 증가함에 따라 사람이 아닌 기계에서 소비되는 영상 정보를 전송하는 요구가 발생하고 있다. 일반적으로 영상 정보를 전송할 때는 전송 비용을 고려하여 정보를 압축하며 기존 영상 압축 방법은 사람의 시각 인지적 특성을 반영하여 설계되었다. 따라서 기존 영상 압축 방법은 기계에서 소비되는 영상 정보를 압축하는 방법으로 적절하지 않다고 판단하여 2019년 7월, 기계를 위한 영상 부호화 기술의 표준화가 시작되었다. 본 논문에서는 머신 비전 태스크 중, 객체 탐지를 수행하는 네트워크의 피처 맵을 압축하는 방법을 제안한다. 제안하는 방법은 피처 맵의 채널 간 중복성을 제거하기 위해 PCA 기반의 변환을 적용하여 피처 맵의 차원을 축소하며 특히 해상도 계층 구조를 갖는 네트워크의 피처 맵을 압축하기 위해 각 해상도 계층간 변환 기저를 예측하여 추가로 압축률을 높인다. 제안하는 방법을 적용하여 객체 탐지 결과의 큰 성능 하락 없이 약 92.3%에 데이터양 감소를 달성하였다.

• Journal of Digital Convergence
• /
• v.11 no.12
• /
• pp.295-302
• /
• 2013

Many reported methods assume that the people in an image or an image sequence have been identified and localization. People detection is one of very important variable to affect for the system's performance as the basis technology about the detection of other objects and interacting with people and computers, motion recognition. In this paper, we present an efficient linear discriminant for multi-view people detection. Our approaches are based on linear discriminant. We define training data with fisher Linear discriminant to efficient learning method. People detection is considerably difficult because it will be influenced by poses of people and changes in illumination. This idea can solve the multi-view scale and people detection problem quickly and efficiently, which fits for detecting people automatically. In this paper, we extract people using fisher linear discriminant that is hierarchical models invariant pose and background. We estimation the pose in detected people. The purpose of this paper is to classify people and non-people using fisher linear discriminant.

• 한국HCI학회:학술대회논문집
• /
• 2006.02a
• /
• pp.1188-1193
• /
• 2006

본 논문에서는 동적 객체의 3 차원 정보를 표현하는 깊이 영상의 노이즈 필터링 방법을 제안한다. 실제 객체의 동적인 3 차원 정보는 적외선 깊이 센서가 장착된 깊이 비디오 카메라를 이용하여 실시간으로 획득되며, 일련의 깊이 영상, 즉 깊이 비디오(depth video)로 표현될 수 있다. 하지만 측정환경의 조명조건, 객체의 반사속성, 카메라의 시스템 오차 등으로 인해 깊이 영상에는 고주파 성분의 노이즈가 발생하게 된다. 이를 효과적으로 제거하기 위해 깊이 영상기반의 모델링 기법(depth image-based modeling)을 이용한 3 차원 메쉬 모델링을 수행한다. 생성된 3 차원 메쉬 모델은 깊이 영상의 노이즈로 인해 경계 영역과 형상 내부 영역에 심각한 형상 오차를 가진다. 경계 영역의 오차를 제거하기 위해 깊이 영상으로부터 경계 영역을 추출하고, 가까운 순서로 정렬한 후 angular deviation 을 이용하여 불필요하게 중복된 점들을 제거한다. 그리고 나서 2 차원 가우시안 스무딩 기법을 적용하여 부드러운 경계영역을 생성한다. 형상 내부에 대해서는 경계영역에 제약조건을 주고 3 차원 가우시안 스무딩 기법을 적용하여 전체적으로 부드러운 형상을 생성한다. 최종적으로 스무딩된 3 차원 메쉬모델을 렌더링할 때, 깊이 버퍼에 있는 정규화된 깊이 값들을 추출하여 원래 깊이 영상과 동일한 깊이 영역을 가지도록 저장함으로서 전역적으로 연속적이면서 부드러운 깊이 영상을 생성할 수 있다. 제안된 방법에 의해 노이즈가 제거된 깊이 영상을 이용하여 고품질의 영상기반 렌더링이나 깊이 비디오 기반의 햅틱 렌더링에 적용할 수 있다.

• The KIPS Transactions:PartD
• /
• v.8D no.4
• /
• pp.367-376
• /
• 2001

For the electronic map service by wireless, the existing map database cannot be used directly. This is because, the data volume of a map is too big to transfer by wireless and although the map is transferred successfully, the devices to display the map usually don’t have enough resources as the ones for desktop computers. It is also not acceptable to construct map database for the exclusive use of wireless service because of the vast cost. We propose new technique to generate a map for wireless service dynamically, from the existing map database. This technique includes the generalization method to reduce the map data volume and filtering method to guarantee that the data volume don’t exceed the limit of bandwidth. The generalization is performed in 3 steps :ㅁ step of merging the layers, a step of reducing the size of spatial objects, and a step of processing user interface. The filtering is performed by 2 module, counter and selector module. The counter module checks whether the data blume of generated map by generalization, exceeds the bandwidth limit. The selector module eliminates the excess objects and selects the rest, on the basis of distance.

• The Journal of the Korea Contents Association
• /
• v.8 no.1
• /
• pp.203-211
• /
• 2008

Due to the advent of presumer and digital production of moving picture, a lot of UCC has been generated by editing the previous moving pictures in various way. It causes a lot of problems on copyright, duplication, and content. In reorganizing the contents, it is necessary to increase productivity and reusability by managing production pipeline systematically through the standardization of moving picture content. For this purpose, we try to develop the moving picture content management system that can manage all kinds of information on the production pipeline, based on SCORM of e-learning by considering production, publication and re-editing. Using the meta-data of content object, user and producer can easily search and reuse the contents. Hence, they can choose the contents object according to their preference and reproduce their own creative UCC by reorganizing and packaging the selected objects. The management of copyright by the unit of scene would solve the problem of copyright. The sequencing technique of SCORM as an interactive storytelling method makes it possible to produce individual contents by user's preference.

• The KIPS Transactions:PartD
• /
• v.12D no.1 s.97
• /
• pp.73-80
• /
• 2005

The purpose of using CBD in the object-oriented modeling is to improve the software capability by reducing iterative time and space complexity. Despites many achievements of CBD, it is needed to study about design patterns and it's standardization for the increment of CBD design reusability. However, it is rather possible that impetuous constructing meta-pattern languages and pattern repositories make adapting patterns to software development more complicate and difficult. By applying GoF design patterns to the design of the TMDL(Total Maximum Daily Load) environmental software discipline, this study suggests a method which specifies Pattern names at class names for retrieving, exploring the adapted patterns on the stage of software design without meta-pattern language which is a redundant abstraction, nor additional pattern repositories. Thus, this study can contribute on the reducing iterations and repetitions that are frequently occurred in the process of the environmental software developments.

• The Journal of the Korea Contents Association
• /
• v.6 no.1
• /
• pp.114-125
• /
• 2006

Since wireless Internet services and small mobile communication devices come into wide use as well as the use of GPS is rapidly growing, researches on moving object, whose location information shifts sequently in accordance with time interval, are being carried out actively. Especially, the researches on vehicle moving object are applied to Advanced traveler information system, vehicle tracking system, and distribution transport system. These systems are very useful in searching previous positions, predicted future positions, the optimum course, and the shortest course of a vehicle by managing historical data of the vehicle movement. In addition, vehicle historical data are used for distribution transport plan and vehicle allocation. Vehicle historical data are stored at regular intervals, which can have a pattern. For example, a vehicle going repeatedly around a specific section follows a route very similar to another. If historical data of the vehicle with a repeated route course are stored at regular intervals, many redundant data occur, which result in much waste of storage. Therefore this thesis suggest a vehicle historical data store scheme for vehicles with a repeated route course using similar trajectory which efficiently store vehicle historical data.

• Journal of KIISE:Computing Practices and Letters
• /
• v.7 no.4
• /
• pp.299-308
• /
• 2001

The company invests its time and money for temporary maintenance to satisfy the fast change of the computer use environment and the user's demands. Therefore, various problems occur including low performance because of duplication of codes and unstable structures from the restructuring and redevelopment. Furthermore, if a developer, who did not participate in initial process of development, wrote additional program codes to upgrade or restructure, it would cause many problems such as lack or lose of development documentation, understanding of documentation and reuse of existing program language. This study, Object-oriented Refactoring Process, suggests that the developers can reuse object unit to overcome the limit of the reusing code. In addition, this paper shows case study to verify our process by adjusting the project called "D2D", which is a case tool for developing windows system from Company D. Our works get positive results such as improvement of system pelfonnance, decreased cost of development and maintenance and optimizing structure and class.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.2
• /
• pp.111-116
• /
• 2021

This study is a study on improving the pedestrian Care system, which delivers jaywalking events in real time to the autonomous driving control center and Autonomous driving vehicles in operation and issues warnings and announcements to pedestrians based on pedestrian signals. In order to secure reliability of object detection method of pedestrian Care system, the inspection method combined with camera sensor with Lidar sensor and the improved system algorithm were presented. In addition, for the occurrence events of Lidar sensors and intelligent CCTV received during the operation of autonomous driving vehicles, the system algorithm for the elimination of overlapping events and the improvement of accuracy of the same time, place, and object was presented.