• The Transactions of the Korea Information Processing Society
• /
• v.7 no.3
• /
• pp.971-976
• /
• 2000

In this paper a new method for head gesture recognition is proposed. A the first stage, face image data are transformed into low dimensional vectors by principal component analysis (PCA), which utilizes the high correlation between face pose images. The a self organization map(SM) is trained by the transformed face vectors, in such a that the nodes at similar locations respond to similar poses. A sequence of poses which comprises each model gesture goes through PCA and SOM, and the result is stored in the database. At the recognition stage any sequence of frames goes through the PCA and SOM, and the result is compared with the model gesture stored in the database. To improve robustness of classification, probabilistic relaxation labeling(PRL) is used, which utilizes the contextural information imbedded in the adjacent poses.

• Proceedings of the KIEE Conference
• /
• 2000.11b
• /
• pp.277-279
• /
• 2000

This paper proposes two kinds of the eddy current brake which uses permanent magnet. The one, like multipolar excitation consists of hexahedron shape of a segmented permanent magnetic and iron situated in the air-gap. The other, like multipolar excitation consists of only a segmented permanent magnetic. We use a finite element method to compute the flux distribution in the model. Also, we use the Galerkin-FEM with linear interpolation function may oscillate between the adjacent nodes to calculate the braking and attraction force. The advantages of the Halbach array are discussed.

• Journal of Chest Surgery
• /
• v.32 no.12
• /
• pp.1144-1147
• /
• 1999

Primary endobronchial leiomyosarcoma is a very uncommon tumor and its diagnosis by spontaneous expectoration of tumor fragment has never been reported in the literature. We report a patient with primary endobronchial leiomyosarcoma that was diagnosed by spontaneous expectoration of partial tumor tissue. The expectorated tissue was found to be pathologically consistent with leiomyosarcoma. Right lower lobectomy and all lymph node dissections were performed during the operation. Pathologic examination showed that the tumor was histologically identical to the patient's previous expectorated tissue and was confined to the bronchus and did not invade the adjacent pulmonary parenchyma and lymph nodes.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06c
• /
• pp.131-133
• /
• 2012

오늘날 세계적으로 많은 사용자들이 온라인 소셜 네트워크 서비스(online social network service)를 이용하고 있다. 소셜 네트워크 서비스에서 사용자들은 자신에 대한 정보를 지속적으로 업로드하고 갱신하기 때문에, 기존의 웹사이트에 비해 개개인의 프라이버시를 위협할 수 있는 소지가 더 크다. 본 연구는 사용자 본인의 프로필뿐만 아니라 주변 사용자의 정보에 의해 민감한 정보가 유출될 가능성을 증명하고, 이를 사전에 탐지하는 기법을 제안한다. 또한 실제 사용자에 대한 실험을 통해 제안 기법이 민감한 정보를 사전에 탐지해내는데 효과적임을 보인다.

• International Journal of Contents
• /
• v.14 no.4
• /
• pp.57-64
• /
• 2018

Among many UNESCO world heritage sites in Korea, "Historic Village: Hahoe" is adjacent to Nakdong River and it is imperative to monitor the water level near the village in a bid to forecast floods and prevent disasters resulting from floods.. In this paper, we propose a recurrent neural network with multiple hidden layers to predict the water level near the village. For training purposes on the proposed model, we adopt the sixth-order error function to improve learning for rare events as well as to prevent overspecialization to abundant events. Multiple hidden layers with recurrent and crosstalk links are helpful in acquiring the time dynamics of the relationship between rainfalls and water levels. In addition, we chose hidden nodes with linear rectifier activation functions for training on multiple hidden layers. Through simulations, we verified that the proposed model precisely predicts the water level with high peaks during the rainy season and attains better performance than the conventional multi-layer perceptron.

• Structural Engineering and Mechanics
• /
• v.91 no.3
• /
• pp.239-250
• /
• 2024

With the aim to provide better predication about fracture behavior, a numerical simulating strategy based on the rigid spring model is proposed for reinforced concrete (RC) structures in this study. According to the proposed strategy, concrete is partitioned into a series of irregular rigid blocks based on the Voronoi diagram, which are connected by interface springs. Steel bars are simulated by bar elements, and the bond slip element is defined at bar element nodes to describe the interaction between reinforcement and concrete. A concrete damage evolution model based on the separation criterion is adopted to describe the weakening process of interface spring between adjacent blocks, while a nonlinear bond slip model is introduced to simulate the synergy behaviour of reinforced steel bars and concrete. In the damage evolution model of concrete, the influence of compressive stress perpendicular to the interface on the shear strength is considered. To check the effectiveness and applicability of the proposed modelling, experimental and numerical studies about a simply-supported RC beam and the two-notched concrete plates in Nooru-Mohamed's experiment are conducted, and the grid sensitivity are investigated.

• Journal of Korea Multimedia Society
• /
• v.10 no.10
• /
• pp.1347-1355
• /
• 2007

A sensor network is composed of a large number of tiny devices, scattered and deployed in a specified regions. Each sensing device has processing and wireless communication capabilities, which enable it to gather information from the sensing area and to transfer report messages to a base station. The energy-efficient routing paths are established when the base station requests a query, since each node has several characteristics such as low-power, constrained energy, and limited capacity. The established paths are recovered while minimizing the total transmit energy and maximizing the network lifetime when the paths are broken. In this paper, we propose a routing algorithm that each sensor node reports its adjacent link information to the sink node when a sink node broadcasts a query. The sink node manages the total topology and establishes routing paths. This algorithm has a benefit to find an alternative path by reducing the negotiating messages for establishing paths when the established paths are broken. To reduce the overhead of collection information, each node has a link information before reporting to the sink. Because the node recognizes which nodes are adjacent. The proposed algorithm reduces the number of required messages, because sensor nodes receive and report routing messages for establishment at the beginning of configuring routing paths, since each node keeps topology information to establish a routing path, which is useful to report sensing tasks in monitoring environments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.7
• /
• pp.2414-2434
• /
• 2015

Among numerous localization schemes proposed specifically for Wireless Sensor Network (WSN), the range-free localization algorithms based on the received signal strength indication (RSSI) have attracted considerable research interest for their simplicity and low cost. As a typical range-free algorithm, Approximate Point In Triangulation test (APIT) suffers from significant estimation errors due to its theoretical defects and RSSI inaccuracy. To address these problems, a novel localization method called FIAPIT, which is a combination of an improved APIT (IAPIT) and a fuzzy logic system, is proposed. The proposed IAPIT addresses the theoretical defects of APIT in near (it's defined as a point adjacent to a sensor is closer to three vertexes of a triangle area where the sensor resides simultaneously) and far (the opposite case of the near case) cases partly. To compensate for negative effects of RSSI inaccuracy, a fuzzy system, whose logic inference is based on IAPIT, is applied. Finally, the sensor's coordinates are estimated as the weighted average of centers of gravity (COGs) of triangles' intersection areas. Each COG has a different weight inferred by FIAPIT. Numerical simulations were performed to compare four algorithms with varying system parameters. The results show that IAPIT corrects the defects of APIT when adjacent nodes are enough, and FIAPIT is better than others when RSSI is inaccuracy.

• Tuberculosis and Respiratory Diseases
• /
• v.62 no.6
• /
• pp.531-535
• /
• 2007

The esophagus is a rate site for rarely involved site of tuberculosis. The most common cause of esophageal tuberculosis is secondary involvement from adjacent tuberculous lymphadenitis. Esophago-nodal or esophagobronchial fistulas may be formed when tuberculous lymph nodes erode the adjacent esophageal or bronchial wall. We report a patient diagnosed with esophageal tuberculosis, which was complicated by an esophago-mediastinal fistula, by endoscopy, sputum acid fast bacilli (AFB) stain, chest computed tomography (CT), and an esophagogram. The patient was treated with antituberculous agents and chest CT and endoscopy showed that the fistula had closed completely.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.10
• /
• pp.1881-1886
• /
• 2016

In this paper, by using D2D communications that perform direct communications among devices within small transmission range of each device without base station, we assume that a device generates and transmits data packets to other proximate devices and the devices which receive the data packets relay those to other adjacent devices. To maximize the total number of devices which successfully receive data packets, Epidemic routing protocol is considered in this paper. In Epidemic routing protocol, all devices which received data packets try to relay the packets to other adjacent devices. We assume various network environment where devices are densely distributed in specific area(crowded area). In this environment, D2D SD can be a source node and D2D devices can be relay nodes. By setting transmission range of D2D SD and D2D devices as parameters, we analyze performance results of coverage probability of Epidemic routing protocol through intensive simulations.