• Title/Summary/Keyword: Sen의 방법

Search Result 38, Processing Time 0.029 seconds

Arduino-based Heart Rate Device for Smart Healthcare (스마트 헬스케어를 위한 아두이노 기반의 심박 측정기 제작)

  • Shin, Chae-lynn;Cho, Young-bok
    • Proceedings of the Korean Institute of Information and Commucation Sciences Conference
    • /
    • 2021.10a
    • /
    • pp.104-105
    • /
    • 2021
  • Along with the rapid development of ICT, health management is continuing based on various smart healthcare devices in an aging society and non-face-to-face era caused by COVID-19. In this paper, a heart rate monitor was manufactured using the DF Robot SEN0203 heart rate sensor based on Arduino so that you can check your health and maintain a healthy life by measuring the heart rate per minute in real time to support personalized health management. The device manufactured through this thesis can be used in various ways, such as a wristband or a smart watch.

  • PDF

A Log-Energy Feature Normalization Method Using ARMA Filter (ARMA 필터를 이용한 로그 에너지 특징의 정규화 방법)

  • Shen, Guang-Hu;Jung, Ho-Youl;Chung, Hyun-Yeol
    • Journal of Korea Multimedia Society
    • /
    • v.11 no.10
    • /
    • pp.1325-1337
    • /
    • 2008
  • The difference of environments between training and recognition is the major reason of degradation of speech recognition. To solve this mismatch of environments, various noise processing methods have been studied. Among them, ERN(log-Energy dynamic Range Normalization) and SEN(Silence Energy Normalization) for normalization of log energy features show better performance than others. However, these methods have a problem that they can hardly achieve normalization for the relatively higher values of log energy features and the environmental mismatch caused by this problem becomes bigger especially in low SNR environments. To solve these problems, we propose applying ARMA filter as post-processing for smoothing log energy features by calculating the moving average in auto-regression scheme. From the recognition results conducted on Aurora 2.0 DB, the proposed method shows improved recognition results comparing with conventional methods.

  • PDF

Analysis of Water Quality Fluctuations in Upstream Namhan (or South Han) River Watershed using Long-term Statistical Analysis (통계적 경향 분석을 통한 남한강 상류 수계 수질 변동 해석)

  • Byeon, Sangdon;Noh, Yeonjung;Lim, Kyeongjae;Kim, Jonggun;Hong, Eunmi
    • Proceedings of the Korea Water Resources Association Conference
    • /
    • 2020.06a
    • /
    • pp.141-141
    • /
    • 2020
  • 강원도는 산림 비중이 크고 급경사지가 많은 지형적 특성과 여름철 비교적 선선하고 강우량이 많은 기후적 특성 때문에 고랭지 농업이 활발히 이루어지고 있다. 하지만, 많은 부분을 차지하는 고랭지밭 면적 탓에 강우시에 토양침식과 비점오염물이 하천으로 유입되고 있다. 강원도 송천 상류에 위치한 도암호는 장기간 탁수를 저류하고 방류로 인한 해체 논란까지 일으키기도 했다. 이에 정부는 흙탕물 저감과 비점오염원 관리를 위한 국고보조사업과 다양한 환경 정책들을 시행중이다. 효율적이고 효과적인 수질 정책을 시행하고 분석하기 위해서는 장기간의 모니터링 자료를 이용한 통계적 분석을 활용하는 것이 중요하다. 수질 자료는 변동이 심하고, 비정규분포를 이루며 결측치와 검출한계 이하의 값들이 많아 비모수 통계 방법을 널리 사용되어 왔다. 그중에서도 계절적 특성을 갖는 수질자료의 장기경향분석에 적합한 Seasonal Mann-Kendall Test을 사용하여 남한강상류 유역의 수질 경향성을 분석하였으며, Sen's Slope를 구하여 수질 자료의 경향 크기를 구하였다. 하지만, Seasonal Mann-Kendall Test는 연구 기간동안의 경향성을 반영할 수 없다는 단점이 있기 때문에 LOWESS Test를 통해 장기간 수질 자료 사이의 경향성을 분석하였다. 이러한 수질자료의 경향 분석 결과는 유역 내 취약 지점을 확인할 수 있으며, 환경 정책의 효과를 평가하고 보완할 수 있는 자료로 이용될 수 있을 것이다.

  • PDF

Analysis of long-term water level change of Dongrae hot spring using time series methods (시계열 방법을 이용한 동래온천 수위의 장기적인 변화 분석)

  • Jeon, Hang-Tak;Hamm, Se-Yeong;Cheong, Jae-Yeol;Lee, Cheol-Woo;Lee, Jong-Tae;Lim, Woo-Ri
    • Journal of the Geological Society of Korea
    • /
    • v.54 no.5
    • /
    • pp.529-544
    • /
    • 2018
  • Dongrae hot spring belongs to the residual magma type and has a long history of bathing since the Silla dynasty in Korea. Due to long development of hot spring water, it is expected that the amount of hot spring water in Dongrae hot spring has been changed. In this study, long-trem water level data of Dongrae hot spring were examined for recognizing the change of the hot spring. By the fluctuation analysis of the hot spring water level from January 1992 to July 2018, the maximum and minimum annual drawdowns of no. 27 well were 137.70 and 71.60 meters, respectively, with an average drawdown of 103.39 m. On the other hand, the maximum and minimum annual drawdowns of no. 29 well were 137.80 and 71.70 meters, with an average drawdown of 103.49 m. Besides, drawdown rate became bigger in recent years. As a result of analyzing autocorrelation of the two wells, the correlation coefficient ranged from 0.919 to 0.991, showing seasonal groundwater level fluctuation. The cross correlation analysis between water level and precipitation as well as water level and hot spring discharge resulted in the correlation coefficients of -0.280 ~ 0.256 and 0.428 ~ 0.553, respectively. Therefore, using Dongnae hot-spring water level data from 1992 to 2018, the Mann-Kendall test and Sen's test showed that the continuous decline of water level was mainly caused by the pumping of the hot spring water among various reasons.

Spam Text Filtering by Using Sen2Vec and Feedforward Neural Network (문장 벡터와 전방향 신경망을 이용한 스팸 문자 필터링)

  • Lee, Hyun-Young;Kang, Seung-Shik
    • Annual Conference on Human and Language Technology
    • /
    • 2017.10a
    • /
    • pp.255-259
    • /
    • 2017
  • 스팸 문자 메시지를 표현하는 한국어의 단어 구성이나 패턴은 점점 더 지능화되고 다양해지고 있다. 본 논문에서는 이러한 한국어 문자 메시지에 대해 단어 임베딩 기법으로 문장 벡터를 구성하여 인공신경망의 일종인 전방향 신경망(Feedforward Neural Network)을 이용한 스팸 문자 메시지 필터링 방법을 제안한다. 전방향 신경망을 이용한 방법의 성능을 평가하기 위하여 기존의 스팸 문자 메시지 필터링에 보편적으로 사용되고 있는 SVM light를 이용한 스팸 문자 메시지 필터링의 정확도를 비교하였다. 학습 및 성능 평가를 위하여 약 10만 개의 SMS 문자 데이터로 학습을 진행하였고, 약 1만 개의 실험 데이터에 대하여 스팸 문자 필터링의 정확도를 평가하였다.

  • PDF

Spam Text Filtering by Using Sen2Vec and Feedforward Neural Network (문장 벡터와 전방향 신경망을 이용한 스팸 문자 필터링)

  • Lee, Hyun-Young;Kang, Seung-Shik
    • 한국어정보학회:학술대회논문집
    • /
    • 2017.10a
    • /
    • pp.255-259
    • /
    • 2017
  • 스팸 문자 메시지를 표현하는 한국어의 단어 구성이나 패턴은 점점 더 지능화되고 다양해지고 있다. 본 논문에서는 이러한 한국어 문자 메시지에 대해 단어 임베딩 기법으로 문장 벡터를 구성하여 인공신경망의 일종인 전방향 신경망(Feedforward Neural Network)을 이용한 스팸 문자 메시지 필터링 방법을 제안한다. 전방향 신경망을 이용한 방법의 성능을 평가하기 위하여 기존의 스팸 문자 메시지 필터링에 보편적으로 사용되고 있는 SVM light를 이용한 스팸 문자 메시지 필터링의 정확도를 비교하였다. 학습 및 성능 평가를 위하여 약 10만 개의 SMS 문자 데이터로 학습을 진행하였고, 약 1만 개의 실험 데이터에 대하여 스팸 문자 필터링의 정확도를 평가하였다.

  • PDF

A Study on the Personality Structure of Characters in the Spiriting Away of Sen and Chihiro (<센과 치히로의 행방불명>에 나타난 캐릭터 성격구조의 이해)

  • Yang Se-Hyeok;Choa Eun-Jung
    • The Journal of the Korea Contents Association
    • /
    • v.5 no.2
    • /
    • pp.123-136
    • /
    • 2005
  • The personality structure of animation characters is revealed as the characters' roles and behavioral patterns in the drama work as elements of sympathy with the audience. Therefore, character creation is not simply limited to the invention of outward appearance. It must start from understanding the importance of personality structure that leads the plot. Thus the present study examined various personality structures retained in characters, centering on The Spiriting Away Of Sen And Chihiro by director Hayao Miyazaki, who has created unique characters. As parts of this research, first, the general characteristics of animation characters were defined. Second, personality types were examined from Carl G. Jung's psychological viewpoint. Analysis was made by connecting the method of classifying personality types based on the multiplicity of personality, namely, 'extroversion and introversion, intuition and sense, thinking and feeling, and judgment and understanding' to the method of classification based on the organic relations among roles in the drama, namely,'victim, persecutor and savior'. Lastly, the research viewpoint derived from the analysis of characters in The Spiriting Away Of Sen And Chihiro was applied. The present study defined the general characteristic revealed by the personality types of characters and their organic relations as 'the personality structure of characters' and documented data. In addition, this study is regarded as meaning in that it suggested an example of the effective creation of animation characters that win the sympathy of the audience.

  • PDF

Sensitivity Analysis of Hydrodynamic Derivatives on Characteristics of Manoeuvring Motion of Manta-type Unmanned Undersea Test Vehicle (Manta형 무인잠수정의 조종운동 특성에 미치는 유체력미계수의 민감도 해석에 관한 연구)

  • Bae, Jun-Young;Sohn, Kyoung-Ho
    • Journal of Navigation and Port Research
    • /
    • v.32 no.8
    • /
    • pp.603-609
    • /
    • 2008
  • Manta-type Unmanned Undersea Test Vehicle(MUUTV) is based on the same design concept as Unmanned Undersea Vehicle called Manta Test Vehicle(MTV), which was originally built and operated by the Naval Undersea Warfare Center(Lisiewicz et al., 2000, Sirmalis et al. 2001). The authors carried out the sensitivity analysis of the response of manoeuvring motion of MUUTV to changes in hydrodynamic derivatives, In order to calculate the sensitivity indices of hydrodynamic derivatives on MUUTV, the method by Sen(2000) was adopted Basically the dynamic mathematical model with six degrees of freedom by Feldman(1979) is used but a little revised, refered to Sohn et al.(2006) and some experiment in circulating water channel. Through the present research, some hydrodynamic derivatives of significance are found out, and also the numerical simulation using simplified mathematical model based on result of sensitivity analysis is ascertained to be enough for prediction of manoeuvring characteristics of MUUTV.

Dissolved Oxygen Trend in Sapgyo Stream Watershed (삽교천유역의 용존산소 추세)

  • Rim, Chang-Soo
    • Journal of Korea Water Resources Association
    • /
    • v.46 no.6
    • /
    • pp.667-681
    • /
    • 2013
  • In this study, monthly and seasonal dissolved oxygen trends of 19 water quality measurement stations in Sapgyo stream watershed were analyzed using monthly dissolved oxygen (DO) data measured for 16 years (1995~2010). Mann-Kendall trend test and Sen's slope estimator were carried out for trend analysis. Furthermore, Sapgyo stream watershed was divided into four different sections (Sapgyo stream, Muhan stream, Gykgyo stream, and Sapgyo lake) and chi-square test of homogeneity for DO trend was carried out for four different sections. The study results indicated that most of water quality measurement stations showed increasing or non-significant trend of DO on a monthly and seasonal basis. The chi-square test of homogeneity for each water quality measurement station showed the statistical homogeneity in seasonal DO trend; however, the test results showed the statistical non-homogeneity in monthly DO trend for the stations located in the reservoir. Overall, the dissolved oxygen trend in each water quality measurement station showed different patterns depending on the location of each station and season.

Long-term Trend Analysis of Major Tributaries of Nakdong River Using Water Quality Index (수질지수를 이용한 낙동강 주요 지류지천의 장기 경향성 분석)

  • Park, Jaebeom;Kal, Byungseok;Kim, Sanghun
    • Journal of Wetlands Research
    • /
    • v.20 no.3
    • /
    • pp.201-209
    • /
    • 2018
  • In this study, the water quality index was calculated using the water quality monitoring data of the major tributaries of the Nakdong River and long-term trend analysis was performed to identify the tributaries requiring priority management. We used a Real-Time Water Quality Index method implemented by the Ministry of Environment. Linear regression as a parametric method and Mann-Kendall Test and Sen Slope Test as a nonparametric method were applied for the trend analysis. The water quality index of major tributaries except for Migeon2 and Seokyo2 was in the range below Fair grade and there were no significant trends for the rest of the sites except Bukan, Chennae, Hogye, Yongdeok. Therefore, in order to improve the water quality of the main stream, management of the tributaries should be preceded.