• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.37 no.6
• /
• pp.1-8
• /
• 2000

This paper describes an algorithm for semantic video object tracking using semi automatic method. In the semi automatic method, a user specifies an object of interest at the first frame and then the specified object is to be tracked in the remaining frames. The proposed algorithm consists of three steps: object boundary projection, uncertain area extraction, and boundary refinement. The object boundary is projected from the previous frame to the current frame using the motion estimation. And uncertain areas are extracted via two modules: Me error-test and color similarity test. Then, from extracted uncertain areas, the exact object boundary is obtained by boundary refinement. The simulation results show that the proposed video object extraction method provides efficient tracking results for various video sequences compared to the previous methods.

• Advances in Computational Design
• /
• v.4 no.2
• /
• pp.179-196
• /
• 2019

Temporary scaffold systems in the construction site play an important role for ensuring the safety of the workers and for constructing the stable structures. To assemble the scaffold, the pipe scaffolding system, the wedge binding scaffolding system and the particular materials have been utilized. To design the material arrangement of a scaffold, firstly the configuration was determined considering the construction geometry. Then, the strength of the scaffold was confirmed and the quantity of the material was accounted. In this paper, the design method of the temporary scaffold was proposed for intending the semi-automatic procedure. In the proposed design method, the geometric design and the safety requirement were specified by the safety standard and the design flow was followed by the designer's knowledge. The size and the quantities of the materials were calculated by referring to the relation between the scaffold and the constructing structure. In the calculating procedure, three dimensional positions of each scaffold materials were calculated and recorded simultaneously. Then, three dimensional scaffold structural was drawn semi-automatically on the CAD software by using the obtained material sizes, positions and directions. The proposed design method provides us the precise quantities of scaffold materials and enables us to reduce the design effort and the cost estimation processes. In addition, the obtained results can be applied to BIM software after converting to IFC format.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.01a
• /
• pp.325-326
• /
• 2022

최근 무인 항공 기술의 발전과 함께 드론의 상업적 활용이 가속화되고 있다. 산업계 및 학계에서는 인공지능, 사물인터넷 등 지능정보기술을 활용하여 드론 응용에 자율 비행을 적용하고자 노력하고 있지만 해결되지 못한 문제들이 산적해 있다. 그러므로 완전 자율 비행은 아니지만, 드론 조종에 경험이 없는 비전문가도 드론을 용이하게 조정할 수 있는 반자동 방식의 비행 제어를 채택한 드론 응용이 개발이 필요하다. 따라서 본 논문에서는 실내환경에서 정찰 임무 수행을 위한 반자동 비행 조종 시스템을 제안한다. 제안한 시스템은 교차로 지점마다 사용자에게 이동 방향을 선택할 기회를 제공하고 교차로 간 일정한 속도 및 벽면과의 일정한 간격으로 자율비행하도록 제어한다. 또한, 실내 측위 기술인 ORB-SLAM 알고리즘을 이용하여 드론이 실내 공간 및 자신의 위치를 파악하도록 한다.

• Journal of the Korean Society for information Management
• /
• v.24 no.1 s.63
• /
• pp.251-271
• /
• 2007

This paper studies the problem of classifying documents with labeled and unlabeled learning data, especially with regards to using document similarity features. The problem of using unlabeled data is practically important because in many information systems obtaining training labels is expensive, while large quantities of unlabeled documents are readily available. There are two steps In general semi-supervised learning algorithm. First, it trains a classifier using the available labeled documents, and classifies the unlabeled documents. Then, it trains a new classifier using all the training documents which were labeled either manually or automatically. We suggested two types of semi-supervised learning algorithm with regards to using document similarity features. The one is one step semi-supervised learning which is using unlabeled documents only to generate document similarity features. And the other is two step semi-supervised learning which is using unlabeled documents as learning examples as well as similarity features. Experimental results, obtained using support vector machines and naive Bayes classifier, show that we can get improved performance with small labeled and large unlabeled documents then the performance of supervised learning which uses labeled-only data. When considering the efficiency of a classifier system, the one step semi-supervised learning algorithm which is suggested in this study could be a good solution for improving classification performance with unlabeled documents.

• Bulletin of the Korean Mathematical Society
• /
• v.20 no.2
• /
• pp.71-74
• /
• 1983

The problems of the continuity of homomorphisms between Banach algebras have been studied widely for the last two decades to obtain various fruitful results, yet it is far from characterizing the calss of Banach algebras for which each homomorphism from a member of the class into a Banach algebra is conitnuous. For commutative Banach algebras A and B a simple proof shows that every homomorphism .theta. from A into B is continuous provided that B is semi-simple, however, with a non semi-simple Banach algebra B examples of discontinuous homomorphisms from C(K) into B have been constructed by Dales [6] and Esterle [7]. For non commutative Banach algebras the problems of automatic continuity of homomorphisms seem to be much more difficult. Many positive results and open questions related to this subject may be found in [1], [3], [5] and [8], in particular most recent development can be found in the Lecture Note which contains [1]. It is well-known that a$^{*}$-isomorphism from a $C^{*}$-algebra into another $C^{*}$-algebra is an isometry, and an isomorphism of a Banach algebra into a $C^{*}$-algebra with self-adjoint range is continuous. But a$^{*}$-isomorphism from a $C^{*}$-algebra into an involutive Banach algebra is norm increasing [9], and one can not expect each of such isomorphisms to be continuous. In this note we discuss an isomorphism from a commutative $C^{*}$-algebra into a commutative Banach algebra with dense range via separating space. It is shown that such an isomorphism .theta. : A.rarw.B is conitnuous and maps A onto B is B is semi-simple, discontinuous if B is not semi-simple.

• Journal of the Korea Society for Simulation
• /
• v.23 no.4
• /
• pp.171-180
• /
• 2014

Our Automatic Replanning enables simulation entities to execute goal oriented behavior planning by dynamic behavior linking. Existing methods especially in Semi-Automated Forces (SAF) are mainly executing strict plans which are given at initial stage, thus they are not effective to cope with contingencies especially in a human in the loop simulation where humans interrupt. Moreover, those usually suffer from explosion of behavior combination in attempt to describe all possible countermeasures, and such combinations may be prone to being inconsistent to the situations. Our method generates behavior sequence in which behavior are linked from the goal in the manner of back-propagation. Each behavior has tags of pre/post-conditions. The tags are linked dynamically according to a certain contingency. The method is being applied to a national defense research project to show feasibility.

• Journal of Digital Convergence
• /
• v.18 no.8
• /
• pp.505-509
• /
• 2020

A migraine was a headache disorder characterized by recurrent moderate to severe headaches. The diagnosis was based on clinical signs and symptoms. Medication, physical therapy, nerve block, and nerve stimulation could be applied for treatment. This report described a case of severe migraine without aura that lasted several weeks periodically in a 59-year-old woman. Periodic headache had lasted for more than 14 years, and although she took medicines and nerve blocks, severe pain (VAS 7) was persisted. We recommended her to use the thermo-spinal massage device (CGM MB-1401, CERAGEM Inc., Cheonan, South Korea) continuously three times a week applying in semi-automatic mode around the neck for 40 minutes. There was no change in the pain scale in the automatic mode for the first 4 weeks. Subsequently, the semi-automatic mode of the cervical area was treated for 2 weeks to relieve the pain scale, and it was confirmed that the relieved state maintained for 2 months. This case highlighted the importance in considering thermo-spinal massage devices for managing migraine without aura.

• The KIPS Transactions:PartD
• /
• v.13D no.3 s.106
• /
• pp.299-308
• /
• 2006

As recently XML is becoming the standard of exchanging web documents and public documentations, XML data are increasing in many areas. To retrieve the information about XML documents efficiently, the semantic web based on the ontology is appearing. The existing ontology has been constructed manually and it was time and cost consuming. Therefore in this paper, we propose the semi-automatic ontology generation technique using the data mining technique, the association rules. The proposed method solves what type and how many conceptual relationships and determines the ontology domain level for the automatic ontology generation, using the data mining algorithm. Appying the association rules to the XML documents, we intend to find out the conceptual relationships to construct the ontology, finding the frequent patterns of XML tags in the XML documents. Using the conceptual ontology domain level extracted from the data mining, we implemented the semantic web based on the ontology by XML Topic Maps (XTM) and the topic map engine, TM4J.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2000.11b
• /
• pp.489-497
• /
• 2000

Since Chinese cabbages weigh 3 to 5kgf and are big in size at the time of harvest, handling operations such as harvesting, loading and unloading including transportation require the highest labor demand among all other cultivation processes. Recently, though several cabbage harvesters were developed in Japan and Europe, those harvesters were not suitable for Chinese cabbages cultivated in Korea because of the size and shape. The cabbage harvester is almost meaningless without any proper cabbage piling and pallet unloading mechanism. Most harvesters developed so far adopted a sort of slide and free falling way in collecting cabbages into the pallet. Three or four labors are usually required for cleaning incoming cabbages and loading those in the pallet. Because of the required time for piling cabbages without severe damage and the required space capacity to carry empty and loaded pallets, harvesting speed should be adjusted in accordance with time required for consecutive operations. Up to now, any automatic or semi-automatic collecting device has not been developed in the world to pile cabbages on the layer one by one into the pallet in the ordered way with little damage and to unload pallet from the harvester continuously during the harvest process. To compromise system expenses and function, Semi-automatic cabbage piling and pallet unloading mechanism was devised and it required one labor. The foldable mesh pallet with a size of 1050mm x 1050mm x 1000mm and holding capacity of around 70 cabbages was utilized. The prototype for piling and unloading mechanism was composed of three parts such as feeding device, automatic piling device with retractable bellows, and pallet unloading device. Prior to developing the prototype, the geometric properties and the amount of the damage of the cabbage caused during the piling operation were investigated. Considering the height of the pallet, a series of cabbage carrying plates were mounted to the bracket chain to lift and to carry cabbages to the loading device. Indoor laboratory experiments showed that the cabbage carrying chain conveyor worked successfully. Considering the conveying speed 0.46m/sec of the pull up belt from the cabbages on the ground, the speed of cabbage carrying chain conveyor worked property in the range of 0.26m/sec to 0.36m/sec. The system allowed the operator to modify the position of cabbage slightly. Overall system worked successfully resulting into almost same capacity without severe damage to the cabbage as human did.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2002.04a
• /
• pp.357-362
• /
• 2002

Most of battery industries are growing explosively as a core strategy industry for the development of the semi-conductor, the LCD, and the mobile communication device. In this thesis, dynamic characteristics of the steel can labeling machine on the automatic cell assembly line are studied. Dynamic characteristic analysis consists of dynamic behavior analysis and finite element analysis and is necessary for effective design of machines. In the dynamic behavior analysis, the displacement, velocity, applied force and angular velocity of each components are simulated according to each part. In the FEA, stress analysis, mode analysis, and frequency analysis are performed for each part. The results of these simulations are used for the design specification investigation and compensation for optimal design of cell manufacturing line. Therefore, Virtual Engineering of the steel can labeling machine on the automatic cell assembly line systems are modeled and simulated. 3D motion behavior is visualized under real-operating condition on the computer window. Virtual Prototype make it possible to save time by identifying design problems early in development, cut cost by reducing making hardware prototype, and improve quality by quickly optimizing full-system performance. As the first step of CAE which integrates design, dynamic modeling using ADAMS and FEM analysis using NASTRAN are developed.