http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
An FPGA-based Parallel Hardware Architecture for Real-time Eye Detection
Dongkyun Kim,Junhee Jung,Thuy Tuong Nguyen,Daijin Kim,Munsang Kim,Key Ho Kwon,Jae Wook Jeon 대한전자공학회 2012 Journal of semiconductor technology and science Vol.12 No.2
Eye detection is widely used in applications, such as face recognition, driver behavior analysis, and human-computer interaction. However, it is difficult to achieve real-time performance with software-based eye detection in an embedded environment. In this paper, we propose a parallel hardware architecture for real-time eye detection. We use the AdaBoost algorithm with modified census transform(MCT) to detect eyes on a face image. We parallelize part of the algorithm to speed up processing. Several downscaled pyramid images of the eye candidate region are generated in parallel using the input face image. We can detect the left and the right eye simultaneously using these downscaled images. The sequential data processing bottleneck caused by repetitive operation is removed by employing a pipelined parallel architecture. The proposed architecture is designed using Verilog HDL and implemented on a Virtex-5 FPGA for prototyping and evaluation. The proposed system can detect eyes within 0.15 ㎳ in a VGA image.
개방형 가상스위치 기반의 패킷가속화기술을 이용한 네트워크 성능 측정 및 분석
김기현 ( Ki-hyeon Kim ),김용환 ( Yong-hwan Kim ),김주범 ( Joobum Kim ),김동균 ( Dongkyun Kim ) 한국정보처리학회 2017 한국정보처리학회 학술대회논문집 Vol.24 No.1
제 4차 산업혁명의 등장으로 다양한 기술들이 주목받고 있으며, 이 중에서 가장 주목 받고 있는 기술은 빅 데이터 기술이다. 이에 따라 최근 빅 데이터를 이용하기 위한 기업들이 크게 증가하고 있고, 해당 기업들을 위해서 방대한 데이터를 빠르게 전송 및 처리할 수 있는 고성능 네트워킹의 필요성이 증가하고 있다. 데이터의 전송과 처리 속도를 향상시키는 직접적인 방안으로 물리적인 네트워크 장비를 증설할 수 있지만 이는 상당한 비용의 증가를 초래하므로, 이를 해결하기 위해 네트워크 가상화 기술이 대두되었다. 하지만 네트워크 가상화 기술은 네트워크의 성능을 보장할 수 없다는 문제점을 가진다. 이러한 문제가 발생하는 주된 이유는 서버의 운영체제 커널 단에서 패킷을 처리하는 과정에서 성능을 저하시키는 요소들이 다수 존재하기 때문이며, 이를 해결하기 위해 나타난 기술이 패킷가속화기술이다. 본 논문에서는 개방형가상스위치 기반의 패킷가속화기술을 적용한 실험환경을 구성한 후, 이를 통해 가상 스위치 기반의 성능 시험과 패킷가속화기술을 이용한 서비스 체이닝 기술에 대한 성능 시험을 수행했다. 그리고 두 가지 시험을 통해 패킷가속화기술의 안정성과 성능을 검증하였다.
Redundant-Path Scheme for Efficient Interworking between a Wireless Sensor Network and the Internet
Kim, Kyuhyung,Kim, Dongkyun,Kim, Dongwon SAGE Publications 2013 International journal of distributed sensor networ Vol.2013 No.-
<P>A wireless sensor network (WSN) has become an important technology and has been deployed in many emerging applications, including home automation, health care, and precision agriculture. To deploy more service applications, the WSN must solve the internetworking problem (difference in speed, protocol stack, beacon collision etc.). In this study, we propose a redundant-path scheme based on a multichannel WSN for an efficient interworking among the heterogeneous networks. We evaluate the performance of our proposed scheme through a real-world simulation, and the results show an improvement in data throughput, packet delay, and beacon loss ratio.</P>
Performance improvement of TCP in ad hoc networks by mitigating channel contention
Kim, Dongkyun,Yoo, Hongseok John Wiley Sons, Ltd. 2009 WIRELESS COMMUNICATIONS AND MOBILE COMPUTING Vol.9 No.10
<P>In ad hoc networks, the spatial reuse property limits the number of packets which can be spatially transmitted over a path. In standard Transmission Control Protocol (TCP), however, a TCP sender keeps transmitting packets without taking into account this property. This causes heavy contention for the wireless channel, resulting in the performance degradation of TCP flows. Hence, two techniques have been proposed independently in order to reduce the contention. First, a TCP sender utilizes a congestion window limit (CWL), by considering the spatial reuse property. This prevents the TCP sender from transmitting more than CWL number of packets at one time. Second, a delayed ack (DA) strategy is exploited in order to mitigate the contention between the TCP ACK and DATA packets. Recently, although TCP-DAA (Dynamic Adaptive Acknowledgment) attempts to utilize a CWL-based DA strategy, TCP-DAA overlooks a dynamic correlation between these two techniques. This paper, therefore, reveals the dynamic correlation and also proposes a protocol which not only reduces the frequency of the TCP ACK transmissions but also determines a CWL value dynamically, according to network conditions. Simulation studies show that our protocol performs the best in various scenarios, as compared to TCP-DAA and standard TCP (such as TCP-NewReno). Copyright © 2009 John Wiley & Sons, Ltd.</P>
Kim Dongkyun,Oh Jaehoon,Im Heeju,Yoon Myeongseong,Park Jiwoo,Lee Joohyun 대한의학회 2021 Journal of Korean medical science Vol.36 No.27
Background: Rapid triage reduces the patients' stay time at an emergency department (ED). The Korean Triage Acuity Scale (KTAS) is mandatorily applied at EDs in South Korea. For rapid triage, we studied machine learning-based triage systems composed of a speech recognition model and natural language processing-based classification. Methods: We simulated 762 triage cases that consisted of 18 classes with six types of the main symptom (chest pain, dyspnea, fever, stroke, abdominal pain, and headache) and three levels of KTAS. In addition, we recorded conversations between emergency patients and clinicians during the simulation. We used speech recognition models to transcribe the conversation. Bidirectional Encoder Representation from Transformers (BERT), support vector machine (SVM), random forest (RF), and k-nearest neighbors (KNN) were used for KTAS and symptom classification. Additionally, we evaluated the Shapley Additive exPlanations (SHAP) values of features to interpret the classifiers. Results: The character error rate of the speech recognition model was reduced to 25.21% through transfer learning. With auto-transcribed scripts, support vector machine (area under the receiver operating characteristic curve [AUROC], 0.86; 95% confidence interval [CI], 0.81–0.9), KNN (AUROC, 0.89; 95% CI, 0.85–0.93), RF (AUROC, 0.86; 95% CI, 0.82–0.9) and BERT (AUROC, 0.82; 95% CI, 0.75–0.87) achieved excellent classification performance. Based on SHAP, we found “stress”, “pain score point”, “fever”, “breath”, “head” and “chest” were the important vocabularies for determining KTAS and symptoms. Conclusion: We demonstrated the potential of an automatic KTAS classification system using speech recognition models, machine learning and BERT-based classifiers.