• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.26 no.3B
• /
• pp.338-346
• /
• 2001

이 논문에서는 잡음에 대해 유연성이 있는 신경망과 차영상법-DCT를 이용한 얼굴인식 알고리즘을 제안한다. 동일환경(조도의 세기, 얼굴에서 카메라까지의 거리)에서 연속적으로 두 개의 영상을 캡쳐했다. 이 때 한 영상은 얼굴을 포함하지 않고 다른 영상은 얼굴을 포함하게 된다. 차영상 방법을 이용하여 두 개의 이미지로부터 얼굴영상과 배경영상을 분리하고 그 다움에 분리된 얼굴영역에서 사각영역을 추출하여 이 영역을 얼굴의 특징영역으로 이용하였다. 이 사각 영역은 눈, 코, 입, 눈썹 등이 포함된다. 다음으로 이 영역에 대해 DWT 연산을 수행한후 특징 백터를 추출하였고, 추출된 특징벡터는 정규화 되어 신경망의 입력벡터로 사용되었다. 시뮬레이션 결과 학습된 얼굴영상에 대해서는 100% 인식률을 보였고 학습되지 않는 얼굴 영상에 대해서는 92.25%의 인식률을 보였다.

• The Transactions of the Korea Information Processing Society
• /
• v.1 no.3
• /
• pp.311-318
• /
• 1994

In this paper, a new stroke extraction method of Chinese character base on the human optical field(the Receptive Field of Cell) is proposed. In processing the feature extraction of the chinese character, needed are more perfect extraction methods for separated informations and its data base. This method can be applied to processing neural cell using conventional feature extraction mechanism in the optical boundary of retina and cerebrum. With this method, its applicability and effectiveness were demonstrated extracting strokes from Chinese character.

• Journal of the Korea Society for Simulation
• /
• v.17 no.4
• /
• pp.41-50
• /
• 2008

In this study, an improved scheme for distance-based registration (DBR) is proposed and its performance is analyzed. In the DBR, when a mobile station (MS) enters a new cell, it calculates the distance between last registered cell and current cell and registers its location if the distance reaches reference distance D. In this study, two-location DBR (TDBR) is proposed to improve the performance of the DBR. In the TDBR, an MS stores not only last registered location area (LA) but also previously registered LA, and then no registration is needed when the MS crosses two LAs stored already. However, since the TDBR may increase paging cost, trade-off is necessary between decreased registration cost and increased paging cost. In this study, the performances of two schemes are analyzed and compared using 2-dimensional random walk mobility model in hexagonal cell configuration. We show that our mathematical analysis is accurate by comparing with simulation. From the numerical results for various circumstances, it is shown that our proposed TDBR outperforms current DBR in most cases.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.1
• /
• pp.39-46
• /
• 1997

In this paper, the PA(PhotoAcoustic) signals excited in metals by Xef, KrF excimer laser pulse were detected by a PZT transducer, and its transforming machanism and directivity patterns were analysed. The laser energy density in irradiation spot divides the PA trasnsorming machanism to be classified into thermoelastic and plasma regime, and the transforming machanisms in two regimes are different from each other. Based on theoretical model, it is predicted that shear wave is greater than longitudinal in the thermoelastic regime and longitudinal is greater than shear wave by reaction force in plasma regime. These predictions were verified through experiments by using of the XeF excimer pulsed laser of 480nm center-wavelength and the KrF excimer pulsed laser of 248nm. Also, for its directivity pattern, an arrival angle of the maximum longitudinal energy was around $60^{\circ}$ and maximum shear energy was around $30^{\circ}$ in the thermoelastic regime, and an arrival angle of maximum longitudinal energy was shown on nomal to the surface and maximum shear energy was represented in about $30^{\circ}$ in plasma regime.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.11
• /
• pp.2127-2134
• /
• 2007

The electronic learning helped a learner to overcome the time restriction by providing mobility, instantly and flexibility but the restriction in connection with space on cable computer remained unsolved. Accordingly, the electronic learning has tendency to change into mobile learning environment which allows a learner to overcome time and spatial restriction. However, these mobile devices have a limitation to awareness of learning contents provided over the realtime video movie due to its small display size. Therefore, this paper suggests a technique according to the following priority: for a real time learning image, extract region of study for region of interest, rescale the real time image to its proper size suitable for the display device, and then make it displayed on a wireless PDA. As a result of the experiment, we reduced the calculating time by sampling the field centering on learning contents adaptively and computing the field best suited for device size of the user effectively.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.93-101
• /
• 2021

In this paper, a vehicle license plate detection scheme is proposed that uses the spatial attention areas to detect accurately the license plates in various real-road situations. First, the previous WPOD-NET was analyzed, and its detection accuracy is evaluated as lower due to the unnecessary noises in the wide detection candidate areas. To resolve this problem, a vehicle license plate detection model is proposed that uses the candidate area of the license plate as a spatial attention areas. And we compared its performance to that of the WPOD-NET, together with the case of using the optimal spatial attention areas using the ground truth data. The experimental results show that the proposed model has about 20% higher detection accuracy than the original WPOD-NET since the proposed scheme uses tight detection candidate areas.

• 한국ITS학회:학술대회논문집
• /
• v.2006 no.10
• /
• pp.286-289
• /
• 2006

• Journal of Korea Multimedia Society
• /
• v.9 no.4
• /
• pp.482-489
• /
• 2006

How to process event message traffic efficiently is important to collaboration among multiple distributed users, and it is a key research issue in the virtual environment fields. AOI model is one of solutions for this problem; which reduces message traffic by restricting event propagation region into an area in which a user is interested. However, in previous systems, it is difficult to adapt to dynamic environment where changes in the number and movement of participants are frequent. This paper proposes a new event management model, called extended AOI model, which is helpful to improve flexibility and efficiency of traditional models, and is able to extend its functions by introducing multi-layered regions. The proposed model can be applied to DVE and MMORPG as an event filtering model.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.07a
• /
• pp.537-538
• /
• 2011

기존 ATSC 시스템의 시간영역 채널 추정은 많은 탭을 사용하고 복잡도가 높지만 OFDM과 같이 보호구간을 이용하여 주파수 영역 등화기를 사용하면 복잡도를 낮추면서 정확한 채널 보상이 가능하다. 본 논문은 ATSC 시스템의 프레임에 보호구간을 삽입하여 주파수 영역 채널 등화를 제안하였다. 또한 DTV수신기의 성능을 확인하기 위해 브라질 채널을 이용하여 컴퓨터 모의실험을 통해 성능을 확인하였다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.7 no.1
• /
• pp.32-41
• /
• 2008

The moving object tracking in vision based observation using video uses difference method between GMM(Gaussian Mixture Model) based background and present image. In the case of racking object using binary image made by threshold, the object is merged not by object information but by Cast-Shadow. This paper proposed the method that eliminates Cast-Shadow using backpropagation Neural Network. The neural network is trained by abstracting feature value form training image of object range in 10-movies and Cast-Shadow range. The method eliminating Cast-Shadow is based on the method distinguishing shadow from binary image, its Performance is better(16.2%, 38.2%, 28.1%, 22.3%, 44.4%) than existing Cast-Shadow elimination algorithm(SNP, SP, DNM1, DNM2, CNCC).