http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
목표물의 고속 탐지 및 인식을 위한 효율적인 신경망 구조
원용관(Weon Yong Kwan),백용창(Baek Yong Chang),이정수(Lee Jeong su) 한국정보처리학회 1997 정보처리학회논문지 Vol.4 No.10
Target detection and recognition problems, in which neural networks are widely used, require translation invariant and real-time processing in addition to the requirements that general pattern recognition problems need. This paper presents a novel architecture that meets the requirements and explains effective methodology to train the network. The proposed neural network is an architectural extension of the shared-weight neural network that is composed of the feature extraction stage followed by the pattern recognition stage. Its feature extraction stage performs correlational operation on the input with a weight kernel, and the entire neural network can be considered a nonlinear correlation filter. Therefore, the output of the proposed neural network is correlational plane with peak values at the location of the target. The architecture of this neural network is suitable for implementing with parallel or distributed computers, and this fact allows the application to the problems which require realtime processing. Net training methodology to overcome the problem caused by unbalance of the number of targets and non-targets is also introduced. To verify the performance, the proposed network is applied to detection and recognition problem of a specific automobile driving around in a parking lot. The results show no false alarms and fast processing enough to track a target that moves as fast as about 190 km per hour.
신경회로망 - 퍼지 논리 이론을 이용한 ATM망에 대한 효율적인 트래픽 제어 기법의 해석
이배호(Lee Bae Ho),한성일(Han Seong Il),원용관(Weon Yong Kwan) 한국정보처리학회 1998 정보처리학회논문지 Vol.5 No.2
B-ISDN network based on ATM technology using statistical multiplexing method supports various multimedia services. Because these multimedia services require both the quality of service and the bandwidth, ATM networks need the efficient traffic and congestion control methods to avoid congestion. Also, it is necssary to use statistical multiplexing method having flexibility for both supporting various services and maximizing the utilization of network resources. In this paper, after we compared and analyzed existing algorithm related to the traffic control methods, then we presented an ATM traffic control mechanism. It is focusing on connection admission control and cell multiplexing methods. Also, we considered the interfaces with other control mechanism. It is focusing on connection admission control and cell multiplexing methods. Also, we considered the interfaces with other control mechanisms such as usage parameter control, bandwidth prediction, and congestion control method. We proposed a novel ATM traffic control mechanism using the neural networks and fuzzy logic theory. We simulated the proposed traffic controller compared with the existing controllers. Simulation results showed that the proposed traffic controller outperform the existing controllers.