http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
拱辰丹의 "麝香 代入 沈香 或 木香"에 관한 古典文獻 硏究
이민석,이종훈,윤태관,이장천,이부균,Lee, Min-Seok,Lee, Jong-Hoon,Yoon, Tae-Kwan,Lee, Jang-Cheon,Lee, Boo-Kyun 대한한의학방제학회 2015 大韓韓醫學方劑學會誌 Vol.23 No.2
Objectives : The object of this paper is to seek and provide evidences for the possibilities of substitutional use of Aquilariae Resinatum Lignum and Aucklandiae Radix for Moschus as an ingredient of Gongjin-dan by clarifying the similarities and differences among these herbs in the classic literatures. Methods : We have found out the features of Moschus, Aquilariae Resinatum Lignum and Aucklandiae Radix by searching Qi, Flavor and Effect of each herb in 『Shennongbencaojing』, 『Zhengleibencao』, 『Bencaogangmu』, 『Dongeuibogam』, 『Bencaobeiyao』, 『Euijongsonik』 and 『Bangyakhappyeon』. And we have compared and analyzed the fomula of Gongjin-dan and other prescriptions in 『Beijiqianjinyaofang』, 『Shengjizonglu』, 『Shiyidexiaofang』, 『Dongeuibogam』, 『Euijongsonik』, 『Euibanghwaltu』 and 『Bangyakhappyeon』. Results : We could find out the similarities and differences in Qi, Flavor and Effect among Moschus, Aquilariae Resinatum Lignum and Aucklandiae Radix. And we could also find the examples of substitutional use of Aquilariae Resinatum Lignum as an ingredient of Gongjin-dan by comparing the formula of Gongjin-dan in the classic literatures as above. And in the other prescription, Aucklandiae Radix was also used instead of Moschus. Conclusions : All of three herbs have pungent in Flavor and warm in Qi. And these herbs also have similar effects in dispersing invading pathogenic factors and smoothening the flow of Qi. And according to 『Bencaobeiyao』, 『Euijongsonik』 and 『Bangyakhappyeon』, Aquilariae Resinatum Lignum and Aucklandiae Radix direct and regulate the Qi in common. The substitutional use of Aquilariae Resinatum Lignum and Aucklandiae Radix instead of Moschus in Gonjin-dan seems to be related to 'the ascent of Water and descent of Fire' effect of Gongjin-dan.
유비쿼터스 환경의 위치 기반 모바일 전자상거래 서비스 통합 구조에 관한 연구
이민석,이훈일,이미영,Lee, Min-Suk,Lee, Hun-Ll,Lee, Mi-Young 한국데이터전략학회 2005 Journal of information technology applications & m Vol.12 No.3
The internet and wireless communication technologies are creating ubiquitous environments in which various services are expected anytime and anywhere. Many hardware facilities have been developed and system structures are suggested for mobile services to realize a ubiquitous computing environment with appropriate quality. But these applications are not designed in the consideration of the general capabilities to perform user's wireless and mobile communication/transaction. Consequently, different needs from users are not sufficiently satisfied yet. In this study, we suggest structure of the emerging network system for mobile commerce that provides users with seamless and ubiquitous environments using location information which exploit context-aware technology.
GIS를 활용한 광주광역시 도시재생방향에 대한 기초연구
이민석,Lee, Min-Seok 한국주거학회 2015 한국주거학회 논문집 Vol.26 No.1
The purpose of this study is to present an appropriate approach to urban regeneration by analyzing Gwangju Metropolitan City. For the study, residential area of Asia Culture Complex was analyzed based on GIS. The areas concerned were categorized and looked into by their Situation based on the analysis of urban infrastructure state. However, as the architectural space within a city are limited, and accurate assessment of them is needed. The residential areas have the following characteristics. They become an aging and dilapidating residential and socially underprivileged-concentrated area. It is necessary to set up a direction for a future-oriented planning for the development of these areas. Along with a decreasing population, it is important to improve the urban architecture environment for the elderly in an aging society. The economic growth of a city if necessary for the future. Therefore, the criteria for the assessment is required. And the subsequent setting-up and planning of a future direction for a city is badly needed. After that, an accurate assessment of a city called for. City space allows for various interpretations. As the rich are getting richer and the poor are getting poorer, a wealth is concentrated on one district or an area.
휴대용 게임기를 이용한 공개 SW 기반 임베디드 소프트웨어 교육 IDE 환경 구성
이민석,Lee, MinSuk 대한임베디드공학회 2012 대한임베디드공학회논문지 Vol.7 No.2
In this paper we describe the integrated development environment for embedded software education. Our environment uses a cheap and widely available portable game console, Nintendo DS-lite. We have integrated open source development tools, and have implemented download shell, remote debugger. The results are all open source, and were used for university class for real embedded software education.
다채널 음향방출 신호에 대한 딥러닝을 통한 3차원 위치 표정
임소영(Soyoung Lim),서호건(Hogeon Seo),이항로(Hang-Lo Lee),김진섭(Jin-Seop Kim) 한국비파괴검사학회 2022 한국비파괴검사학회지 Vol.42 No.1
본 연구에서는 다채널 음향방출 신호로부터 발진 위치 좌표를 예측하는 딥러닝 모델을 활용함에 있어서 손실함수 및 데이터 전처리 조건에 따른 성능을 평가했다. 7 채널의 음향방출 신호를 이미지 형태로 모델의 입력으로 활용하였고, 출력은 X, Y, Z 좌표를 각각 출력하는 경우와 한 번에 출력하는 경우를 고려하였다. 딥러닝 모델로는 EfficientNet B3을 활용했으며, 평가지표는 평균 절대 오차를 채택했다. 손실함수로는 평균제곱오차와 평균절대오차, 시계열 데이터 전처리 기법으로는 스케일링(MinMaxScaler, StandardScaler, MaxAbsScaler), 신호 전처리(Abs, MA, LPF), 샘플링 길이(1024, 512, 342)에 따른 위치 표정 성능을 비교했다. 그 결과, 손실함수로는 평균제곱오차를 사용하는 것이 유리했고, 스케일링으로는 MaxAbsScaler가 효과적이었다. MaxAbsScaler 적용 후, 절대값을 취해준 후 저역통과필터나 이동평균을 적용한 데이터의 단순화는 성능에 악영향을 미쳤다. MaxAbsScaler 적용 후, 512 개만 샘플링 했을 때 더 좋은 성능을 보였고, X, Y, Z의 각 좌표를 개별적으로 추정하는 방식이 최고 성능을 보였다. This study evaluated source localization performance based on the loss function and preprocessing conditions when using a deep learning model that predicts the source coordinates from multi-channel acoustic emission signals. Signals of seven channels were fed as input into a model in the form of an image, and two output cases were considered: outputting X, Y, and Z coordinates one at a time and outputting them at the same time. EfficientNet B3 was used as the model, and the mean absolute error was adopted as the performance index. Mean square error and mean absolute error were used as loss functions. The performance was compared with respect to scaling (MinMaxScaler, StandardScaler, MaxAbsScaler), preprocessing (Abs, MA, LPF), and sampling length (1024, 512, 342). The results show the advantage of using the mean squared error and the effectiveness of MaxAbsScaler. After applying MaxAbsScaler, the simplification of the data, such as taking the absolute value, and applying a low-pass filter or moving average method, had an adverse effect on the performance. The MaxAbsScaler exhibited better performance when 512 samples were input, and each coordinate of X, Y, and Z was separately estimated.