http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
조준희(Jun-Hee Cho),이지영(Ji-Young Lee),이운성(Woon-Sung Lee),홍승준(Seung-Jun Hong),임양근(Lyang-Keun Lim) 한국자동차공학회 2003 한국자동차공학회 춘 추계 학술대회 논문집 Vol.- No.-
The Safe Driving Management System (SDMS) is a comprehensive solution that monitors and stores driving conditions of vehicles, detects dangerous driving situations, and analyses the results in real time. The key component of the SDMS is an algorithm that detects and classifies dangerous driving conditions into representative cases such as sudden acceleration, sudden braking, sudden lane change, and sudden turning. Both simulation and vehicle test have been conducted to develop the algorithm. The SDMS is being tested on buses for its reliability and objectivity. It is expected that the system can contribute to prevention of traffic accidents, reduction of commercial vehicle operation costs and insurance fees.<br/>
하태웅,홍승준,Ha, Tae-Woong,Hong, Seung-Jun 한국자동차안전학회 2020 자동차안전학회지 Vol.12 No.3
This paper has presented the headlight inspection standards and methods of two-wheeled motorcycles considering Korean Motor Vehicle Safety Standards (KMVSS), Korean Motor Vehicle Inspection Standards (KMVI) and the inspection standards of the International Motor Vehicle Inspection Committee (CITA). As a result of analyzing the headlight luminous intensity test result with fixed inspection equipment, 21% does not meet the inspection standard proposed in this study, and 33.3% in mobile inspection equipment does not meet. The average luminous intensity of motorcycle less than 50 cc is 8,340 cd, so all the lightweight small sized motor cycles do not meet the proposed headlight inspection standard.
박인송,김관희,홍승준,Park, In-Song,Kim, Guan-Hee,Hong, Seung-Jun 한국자동차공학회 2007 한국 자동차공학회논문집 Vol.15 No.5
Since 2000, sports utility vehicles(SUVs) occupy about 40% of domestic vehicle sales. As sports utility vehicle sales are increased the probability of crash accident between SUVs and passenger vehicles increases. Generally, SUVs are heavier than passenger vehicles and their drive height and front end stillness are higher than passenger vehicles. Because of these characteristics SUVs cause more severe injury and fatal injury in SUV to passenger vehicle head-on impact. To evaluate SUV's aggessivity to passenger vehicle, we carried out SUV to passenger vehicle head-on crash test. And finally the way how to reduce incompatibility between SUVs and passenger vehicles is suggested.
홍승준,구자윤,권영헌,이건상 한양대학교 이학기술연구소 2000 이학기술연구지 Vol.2 No.
본 논문에서는 화자 식별과 얼굴 인식을 이용하여 대규모 보안 시스템을 개발하였다. 인식 에러를 줄이기 위해 전 처리 화자 식별기로 DHMM(Discrete hidden Markov Model)을 사용하였고, 후 처리기에서는 얼굴 인식을 위해 신경망을 사용하였다. Visual C++MFC을 사용하여 Sound Card와 Digital camera를 제어하였다. 그리고 본 논문에서 개발한 프로그램을 사용하여 실시간으로 실험하였다. 본 논문에서 인식에 사용한 특징 벡터로는 Filter bank energy, LSF와 Cepstrum coeffient이고, 얼굴 인식의 대표 벡터는 2차원 Filter bank energy를 사용하였다. 실험에서 사용된 사람은 학교에서 무작위로 83명을 선택하였다. Test를 위해 2049개의 음성 데이터와 365개의 얼굴 이미지를 사용하였다. 복합 모델로 구성한 본 시스템은 등록화자에 대해 98.8%의 Access rate와 비 등록화자의 99.82%의 deny rate을 나타내었다. In this paper, we developed the large scale security system using speaker identification and face recognition. In order to reduce the recognition error we used DHMM(Discrete Hidden Markov Model) in speaker identification for pre-processor and neural network in face recognition for post-processor. To test the on-line performance of total system we directly controlled the sound card and the digital camera in the environment of Visual C++MFC. And we developed the program to test the performance of the system in real time. We used filter bank energies, LSF and cepstrum coefficients as the feature vectors of speech, and the 2-dimensional filter bank energies extracted in the range of the representative feature on face. In experiments, we randomly selected 83 persons in school, and then chosen 10 persons as an enrolled person and 73 persons as a non-enrolled person. So we tested 2049 speech data and 365 image data. It turned out that the hybrid system would have 98.8% access rates for the enrolled persons and 99.82% deny rates for the non-enrolled persons.
Rhizopus oryzae를 이용한 바이오 디젤 생산
박상무,홍승준,이기영,김동운 全南大學校 觸媒硏究所 2002 觸媒硏究 論文集 Vol.23 No.-
Biodiesel was produced from soybean oil and methanol using fermentation broth or lipase. Production yield was 62% and 53.5% with purified Rhizopus oryzae lipase and Rhizopus oryzae broth respectively. This yield of biodiesel was highest in case of water content of 20%