• 대한전자공학회논문지SP
• /
• 제42권5호
• /
• pp.63-72
• /
• 2005

본 논문에서 삼각대 위에 고정시킨 카메라를 회전시켜 취득한 비디오 시퀀스로부터 파노라마 영상을 구성하는 새로운 방법을 제안한다. 제안한 방법은 비디오 시퀀스 중에서 실제로 사용할 프레임을 자동으로 선택하는 단계와 파노라마 영상을 구성하는 모자이킹 단계로 나누어진다. 파노라마 영상을 구성하는데 필요한 프레임은 연속된 프레임 간의 중첩 비율 정보를 이용한 보간 검색을 통해 선택된다 이 방법은 카메라의 속도에 상관없이 적절한 프레임을 고속으로 검색할 수 있다. 모자이킹 단계에서는 4쌍의 특징점을 사용하는 사영 변환식(projective transform)을 이용한다. 기존의 연구에서는 특징점을 화소 기반의 정보만을 이용하여 추출하였지만, 본 논문에서 제안한 방법은 특징점의 위치까지 고려하여 사영 변환식의 정확도를 높였다. 제안한 방법은 다양한 영상 시퀀스에 적용하여 실험하였고, 그 결과 모든 영상에 대해서 좋은 화질의 파노라마 영상을 생성하였다.

• 정보처리학회논문지C
• /
• 제10C권7호
• /
• pp.885-890
• /
• 2003

코어 기반 트리(CBT：Core Based Tree) 멀티캐스트 라우팅은 인터넷을 위한 멀티캐스트 라우팅 프로토콜이다. CBT는 멀티캐스트 연결을 위해 하나의 공유 트리를 구성하고, 이 공유 트리는 코어라는 센터 노드를 루트로 한다. 코어 기반 트리 멀티캐스트 라우팅에서 코어의 위치 선정은 비용과 성능 측면에 많은 영향을 끼치고, 코어의 배치 방법은 네트워크 토폴로지에 따라 달라진다. 본 논문에서는 간단하고 효율적인 코어 선택 방법을 제안하고자 한다. 제안한 방법은 거리 벡터 정보를 필요로 하며, 추가적으로 표본 상관 계수 결과를 이용하여, 멀티캐스트 그룹에 속해있는 멤버들의 배치상태에 따라 적합한 라우팅 알고리즘으로 최소평균비용을 갖는 노드를 코어로 선정하는 방법, 또는 PIM-SM 방법을 선택하는 것이다. 제안한 방법의 성능은 평균지연, 최대 지연, 총 비용과 같은 세부적인 실험을 통해 기존연구와 비교하였으며, 실험 결과 제안한 코어 선택 방법이 매우 효율적임을 알 수 있었다.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.195-206
• /
• 2022

Crack detection is essential for inspection of existing structures and crack segmentation based on deep learning is a significant solution. However, datasets are usually one of the key issues. When building a new dataset for deep learning, laborious and time-consuming annotation of a large number of crack images is an obstacle. The aim of this study is to develop an approach that can automatically select a small portion of the most informative crack images from a large pool in order to annotate them, not to label all crack images. An active learning method with difficulty learning mechanism for crack segmentation tasks is proposed. Experiments are carried out on a crack image dataset of a steel box girder, which contains 500 images of 320×320 size for training, 100 for validation, and 190 for testing. In active learning experiments, the 500 images for training are acted as unlabeled image. The acquisition function in our method is compared with traditional acquisition functions, i.e., Query-By-Committee (QBC), Entropy, and Core-set. Further, comparisons are made on four common segmentation networks: U-Net, DeepLabV3, Feature Pyramid Network (FPN), and PSPNet. The results show that when training occurs with 200 (40%) of the most informative crack images that are selected by our method, the four segmentation networks can achieve 92%-95% of the obtained performance when training takes place with 500 (100%) crack images. The acquisition function in our method shows more accurate measurements of informativeness for unlabeled crack images compared to the four traditional acquisition functions at most active learning stages. Our method can select the most informative images for annotation from many unlabeled crack images automatically and accurately. Additionally, the dataset built after selecting 40% of all crack images can support crack segmentation networks that perform more than 92% when all the images are used.

• 대한수학교육학회지:수학교육학연구
• /
• 제8권1호
• /
• pp.31-39
• /
• 1998

In this study the types of strategies, the reasonable method of the selection of the problem for the recognition of the strategies and the presentation of the problems based on the types of the strategies are explored. The strategies can be used in elementary, middle or high school mathematics. But it can't be expected that students use the strategies unfamiliar to them, so it is necessary to select the strategies applicable according to the degree of the student's affective and cognitive abilities. And it is desirable to select the problems for the recognition of the strategy such as nonroutine problems and open-ended problems, but more study is needed in these areas.

• ETRI Journal
• /
• 제42권4호
• /
• pp.585-595
• /
• 2020

Integration of multiple communication technologies in a smart grid (SG) enables employing cognitive radio (CR) technology for improving reliability and security with low latency by adaptively and effectively allocating spectral resources. The versatile features of the CR enable the smart meter to select either the unlicensed or the licensed band for transmitting data to the utility company, thus reducing communication outage. Demand response management is regarded as the control unit of the SG that balances the load by regulating the real-time price that benefits both the utility company and consumers. In this study, joint allocation of the transmission power to the smart meter and consumer's demand is formulated as a two stage multi-armed bandit game in which the players select their optimal strategies noncooperatively without having any prior information about the media. Furthermore, based on historical rewards of the player, a real-time pricing adaptation method is proposed. The latter is validated through numerical results.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 추계학술대회 학술발표 논문집
• /
• pp.95-98
• /
• 2001

In this paper, we propose a method how to determine the medical section select that get input the conditions of illness from a dialog-box using a fuzzy logic as a guidance system of the Cyber Hospital. It is to offer an output the most matched section for medical treatment using the Algebraic product fuzzy inference and Max inference after match the membership degree table of symptoms of received input from patient and their information. It will present of the direction for ignored patient which of a medical section select.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 1996년도 춘계공동학술대회논문집; 공군사관학교, 청주; 26-27 Apr. 1996
• /
• pp.310-313
• /
• 1996

Mathematical programming method for finding optimal solution of job shop scheduling is inadequate to real situation because fo too much computation time. In contrast, dispatching rule is helpful for reducing compuation time but is not guaranted to find optimal solution. The purpose of this paper is to develop a new dispatching rule and procedure to minimize mean flow time whose result is near the optimal solution for job shop scheduling. First step is to select machine which have shortest finishing operation time among the schedulable operations. Second step is to select operation with regard to estimated remaining operation time. The suggested rule is compared with nondelay and MWKR rule for three examples, and is confirmed to be most effective to minimize mean flow time.

• International Journal of Automotive Technology
• /
• 제8권3호
• /
• pp.319-325
• /
• 2007

Preload of critical engine bolts affects the performance and durability of engines. In modern engines that pursue higher power outputs and which are of lighter weight, it becomes more difficult to select an optimal target preload in consideration of various factors such as the role and structural characteristics of joint members, joint load, and fatigue durability of bolts and joint members. A procedure to select the bolt preload using computer-aided engineering technology, especially the finite element method, has been developed. The procedure is illustrated with connecting-rod bolts for which an appropriate preload is known. The selection criteria of target preload and the finite element modeling technique for connecting-rod bolts are also explained.

• 대한임베디드공학회논문지
• /
• 제10권3호
• /
• pp.165-170
• /
• 2015

SNOMED CT browser is a browsing tool for searching clinical terms in SNOMED CT which is a standard terminology set used worldwide. The search result view of previous browsers merely list up candidate terminologies. The problem is that most of users become confused about how to select an appropriate term from the list. This leads serious waste of medical recoding cost. This paper discusses characteristics of SNOMED CT dataset and proposes a novel design of enhanced result view by restructuring the results using relationships of SNOMED CT concepts. Using the proposed scheme, medical doctors or officers can select appropriate terms more efficiently and can reduce overall recording time.

• 산업경영시스템학회지
• /
• 제35권4호
• /
• pp.98-109
• /
• 2012

Recently, the optimisation of end-of-life (EOL) product remanufacturing processes has been highlighted. In particular, computer remanufacturing becomes important as the amount of disposed of computers is rapidly increasing. At the computer remanufacturing, depending on the selections of used computer parts, the value of remanufactured computers will be different. Hence, it is important to select appropriate computer parts at the reassembly. To this end, this study deals with a decision making problem to select the best combination of computer parts for minimising the total remanufacturing computer cost. This problem is formulated with an integer nonlinear programming model and heuristic search algorithms are proposed to resolve it.