http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Image Processing by a Diffusion Neural Network
권율,남기곤,윤태훈,김재창,Kwon, Yool,Nam, Ki-Gon,Yoon, Tae-Hoon,Kim, Jae-Chang The Institute of Electronics and Information Engin 1993 전자공학회논문지-B Vol.b30 No.1
A Gaussian is formed by diffusing a spot excitation. In this paper, a diffusion neural network model is derived from the diffusion equation. And it is shown that a difference of two Gaussians(DOG) may have the same shape as a Laplacian of Gaussian(LOG), A neural network model executing a DOG convolution by diffusing an external excitation is proposed. By this model intensity changes of image may be detected. This model may be implemented economically because each neuron has only four fixed-valued synapes.
Detection of Intensity Changes by a Diffusion Neural Network
권율,남기곤,윤태훈,김재창,Kwon, Yool,Nam, Ki-Gon,Yoon, Tae-Hoon,Kim, Jae-Chang The Institute of Electronics and Information Engin 1992 전자공학회논문지-B Vol.b29 No.11
In this paper we propose a diffusion neural network model. In this model, each excitatory and inhibitory neuron has the capability of diffusing external excitations. We show that this model can be used for the detection of intensity changes of an input image. The relations between the diffusion coefficient, the iteration number of diffusion, and the detected spatial frequency are analyzed. The calculation time is reduced than that of a LOG(a Laplacian of a Gaussian) method.
최태완,권율,김재창,남기곤,윤태훈,Choi, Tae-Wan,Kwon, Yool,Kim, Jae-Chang,Nam, Ki-Gon,Yoon, Tae-Hoon 대한전자공학회 1995 전자공학회논문지-B Vol.b32 No.1
The diffusion neural network can be cfficiently applied to the Gaussian processing. For example, a difference of two Gaussians(DOG) is performed by this network with ease. In this paper, we model a neural network to perform the function /t(.del.${\Delta}^{2}$G) by using the diffusion neural network. This model is used to detect the edges of moving target in image. By this model not only moving target is separated from stationary background but also their trajectories are obtained using accumulated past information in the diffusion neural network. Furthermore this model needs a small number of connections per cell and the connection weights are fixed-valued. Therefore its hardware can be easily implemented with simple structure.
확산 신경 회로망을 이용한 광대역 공간 주파수 성분의 윤곽선 검출
이충호,권율,김재창,남기곤,윤태훈,Lee, Choong-Ho,Kwon, Yool,Kim, Jae-Chang,Nam, Ki-Gon,Yoon, Tae-Hoon 대한전자공학회 1995 전자공학회논문지-B Vol.b32 No.1
The diffusion neural network forms a Gaussian distribution by transferring an excitation to the surround. A DOG(difference of two Gaussians) is obtained by the diffusion neural network. This type of the DOG, which can detect the intensity changes of an image, has the same shape as a LOG(Laplacian of a Gaussian:${\Delta}^2$G) and narrow band pass characteristics. In this paper we show that another type of the DOG which has a very narrow Gaussian for the excitatory and a very wide Gaussian for the inhibitory, can be formed by the diffusion process of this network, This type of the DOG has a wide band width in spatial frequency domain and can be used efficiently in detecting special type of edges.
곽동용,한용민,권율,박홍식,Kwak, Dong-Yong,Han, Yong-Min,Kwon, Yool,Park, Hong-Shik 한국통신학회 1996 韓國通信學會論文誌 Vol.21 No.11
본 논문은 쉐이핑 기능에 기인한 셀 지연과 버퍼의 크기를 조정할 수 있는 한개의 임계 값을 가진 셀 간격 조정 알고리즘을 제안하고 리키버킷 알고리즘을 통과할 수 있는 가장 worst한 트래픽을 셀 간격 조정 알고리즘의 입력으로 가정하여 임계값의 위치에 따라 셀 간격 조정 알고리즘을 통과할 수 있는 트래픽 형태를 규정한다. 그리고 이 트래픽들이 스위치의 지연 및 버퍼 크기에 미치는 영향을 대해 기존의 다른 셀 간격 조정 알고리즘과 비교하였다. 그 결과 제안 알고리즘이 임계값이 없는 기존의 알고리즘보다 쉐이핑에 기인한 지연 및 출력 버퍼 크기를 임계 값에 따라 쉽게 조정할 수 있음을 보여 주었다. In this paper we propose a new shaping algorithm which can control the shaping delay and the output buffer size based on the leaky bucket counter with a threshold value. This paper assumes that input traffic of the proposed shaping algorithm is the worst case traffic tolerated by the continuous leaky bucket algorithm and claracterizes traffic patterns that can depart from our shaping algorithm. We also compare shaping delay and output buffer size of the proposed algorithm with the existing shaping algorithm without a threshold value. Our results show that the proposed shaping algorithm can easily manage the shaping delay and output buffer size than any other mechanism.
B-ISDN 망에서 공통선 신호 기능의 구현 및 성능 평가
이우섭,김화숙,안윤영,권율,Rhee, Woo-Seop,Kim, Hwa-Suk,An, Yoon-Young,Kwon, Yool 한국통신학회 1998 韓國通信學會論文誌 Vol.23 No.5
미래의 통신 망을 위한 서비스 망들은 ATM을 기반으로 하는 B-ISDN 망으로 통합될 것이며 이러한 서버스 망들은 사용자 요구 서비스를 제어하기 위한 신호 전달 망으로 공통선 선호 방식을 사용하게 된다. 이에 따라, ITU-T 에서는 기존의 N-ISDN 공통선 신호의 MTP 신호 계층을 대신하여 B-ISDN 선호 계층들이 권고되었다. 본 논문 에서는 B-ISDN 공중망의 ATM 교환 시스탬에 구현된 공통선 선호 기능에 대해 각 선호 계층별 특성, 기능 및 실현 구조등을 제안하고 성능을 분석하였다. SAAL 계층은 linked-list와 단위 프레임 길이를 사용하는 SSCOP 송수신 버퍼 구조를 제안하고 성능을 분석하였으며, MTP-3b 계층에 대해서도 ATM 교환 시스템 구조에 따른 실현 구조 및 내부 라우팅 방법을 제안하고 그 성능을 분석하였다. 또한, B-ISDN 도입 초기에 나타나는 기존의 N-ISDN 망과 B-ISDN의 SS No.7 연동에 대한 효율적인 연동 구조로서 회선 관련 신호망은 대응 모드만을 사용하는 ISUP/B-ISUP 레벨 연동을 제안하였다. Service networks for the future communication networks will be combined by the B-ISDN networks. These service networks also will use SS No.7 as the signaling transport network for the control of user requriement service. Therefore, ITU-T recommended B-ISDN signaling layers for SS No.7 as a substitute for N-ISDN MTP signaling layer. In this paper, we propose the implementation structure and describe the characteristics and functions of each signaling layer of SS No.7, which are adapted to ATM switching system, and evaluate a performance. The structure of SSCOP transmission buffer using a linked list and an unit frame length is proposed for SAAL layer and the implementation structure and internal routing method according to the ATM switching system are also proposed for MTP-3b layer. Additionally, we propose the ISUP/B-ISUP level interworking structure using only associated mode, which are presented in the first stage of B-ISDN as the effective internatworking structure of SS No.7 for the circuit related signaling network between the existing N-ISDN networks and B-ISDN networks.