• Title/Summary/Keyword: General Component Analysis

Search Result 477, Processing Time 0.032 seconds

Robust Feature Parameter for Implementation of Speech Recognizer Using Support Vector Machines (SVM음성인식기 구현을 위한 강인한 특징 파라메터)

  • 김창근;박정원;허강인
    • Journal of the Institute of Electronics Engineers of Korea SP
    • /
    • v.41 no.3
    • /
    • pp.195-200
    • /
    • 2004
  • In this paper we propose effective speech recognizer through two recognition experiments. In general, SVM is classification method which classify two class set by finding voluntary nonlinear boundary in vector space and possesses high classification performance under few training data number. In this paper we compare recognition performance of HMM and SVM at training data number and investigate recognition performance of each feature parameter while changing feature space of MFCC using Independent Component Analysis(ICA) and Principal Component Analysis(PCA). As a result of experiment, recognition performance of SVM is better than 1:.um under few training data number, and feature parameter by ICA showed the highest recognition performance because of superior linear classification.

Face recognition rate comparison using Principal Component Analysis in Wavelet compression image (Wavelet 압축 영상에서 PCA를 이용한 얼굴 인식률 비교)

  • 박장한;남궁재찬
    • Journal of the Institute of Electronics Engineers of Korea CI
    • /
    • v.41 no.5
    • /
    • pp.33-40
    • /
    • 2004
  • In this paper, we constructs face database by using wavelet comparison, and compare face recognition rate by using principle component analysis (Principal Component Analysis : PCA) algorithm. General face recognition method constructs database, and do face recognition by using normalized size. Proposed method changes image of normalized size (92${\times}$112) to 1 step, 2 step, 3 steps to wavelet compression and construct database. Input image did compression by wavelet and a face recognition experiment by PCA algorithm. As well as method that is proposed through an experiment reduces existing face image's information, the processing speed improved. Also, original image of proposed method showed recognition rate about 99.05%, 1 step 99.05%, 2 step 98.93%, 3 steps 98.54%, and showed that is possible to do face recognition constructing face database of large quantity.

A Study on the Factors Affecting the Subjective Symptoms of VDT Syndrome (VDT증후군 자각증상에 영향을 미치는 인자들에 관한 연구)

  • Moon, Jai-Dong;Lee, Min-Chull;Kim, Byong-Woo
    • Journal of Preventive Medicine and Public Health
    • /
    • v.24 no.3 s.35
    • /
    • pp.373-389
    • /
    • 1991
  • In order to develop the measuring tool of VDT syndrome and investigate the variables affecting the development of VDT syndrome, a questionnaire study accompanied with the evaluation of working environment was performed with 138 VDT users from six public organs in Kwangju area. The results were summarized as follows. 1. As a result of analysis with data collected by newly developed questionnaire, VDT syndrome included five factors named as eye-related component, psychological component, general body component, musculoskeletal component, and skin-related component and the estimates of the internal consistency of five factors were 0.877, 0.820, 0.796, 0.791, 0.593 respectively. 2. Variables affecting the level of eye-related symptoms were the type of main b using VDT, the total time of VDT operation per day, and the use of external filter on CRT. 3. The level of eye-related symptoms in the group using external filter was higher significantly than that in the group not using filter. 4. The past history of severe illness affected the level of psychological symptoms significantly. 5. Variables affecting the level of general body symptoms were b satisfaction and income satisfaction. 6. Variables affecting the level of musculoskeletal symptoms were the type of main job using VDT, whether majored in EDPS, the level of typewriting, b satisfaction, and the total time of VDT operation per day. 7. Age and the use of external filter were significantly related to the level of skin-related symptoms.

  • PDF

Affecting Factors on the Variation of Atmospheric Concentration of Polycyclic Aromatic Hydrocarbons in Central London

  • Baek, Sung-Ok;Roger Perry
    • Journal of Korean Society for Atmospheric Environment
    • /
    • v.10 no.E
    • /
    • pp.343-356
    • /
    • 1994
  • In this study, a statistical investigation was carried out for the evaluation of any relationship between polycyclic aromatic hydrocarbons (PAHss) associated with ambient aerosols and other air quality parameters under varying meteorological conditions. Daily measurements for PAHs and air quality/meteorological parameters were selected from a data-base constructed by a comprehensive air monitoring in London during 1985-1987. Correlation coefficients were calculated to examine any significant relationship between the PAHs and other individual variables. Statistical analysis was further Performed for the air quality/meteorological data set using a principal component analysis to derive important factors inherent in the interactions among the variables. A total of six components were identified, representing vehicle emission, photochemical activity/volatilization, space heating, atmospheric humidity, atmospheric stability, and wet deposition. It was found from a stepwise multiple regression analysis that the vehicle emission component is overall the most important factor contributing to the variability of PAHs concentrations at the monitoring site. The photochemical activity/volatilzation component appeared to be also an important factor particularly for the lower molecular weight PAHs. In general, the space heating component was found to be next important factor, while the contributions of other three components to the variance of each PAHs did not appear to be as much important as the first three components in most cases. However, a consistency for these components in their negative correlations with PAHs data was found, indicating their roles in the depletion of PAHs concentrations in the urban atmosphere.

  • PDF

Shape Design of Passages for Turbine Blade Using Design Optimization System (최적화설계시스템을 이용한 터빈블레이드 냉각통로의 형상설계)

  • Jeong Min-Joong;Lee Joon-Seong
    • Transactions of the Korean Society of Mechanical Engineers A
    • /
    • v.29 no.7 s.238
    • /
    • pp.1013-1021
    • /
    • 2005
  • In this paper, we developed an automatic design optimization system for parametric shape optimization of cooling passages inside axial turbine blades. A parallel three-dimensional thermoelasticity finite element analysis code from an open source system was used to perform automatic thermal and stress analysis of different blade configuration. The developed code was connected to an evolutionary optimizer and built in a design optimization system. Using the optimization system, 279 feasible and optimal solutions were searched. It is provided not only one best solution of the searched solutions, but also information of variation structure and correlation of the 279 solutions in function, variable, and real design spaces. To explore design information, it is proposed a new interpretation approach based on evolutionary clustering and principal component analysis. The interpretation approach might be applicable to the increasing demands in the general area of design optimization.

Dynamic analysis for complex structures using the improved component mode method (개선된 콤포넌트 모드법을 이용한 거대구조물의 동적해석)

  • 심재수;박명균
    • Proceedings of the Computational Structural Engineering Institute Conference
    • /
    • 1993.04a
    • /
    • pp.37-44
    • /
    • 1993
  • There are a lot of linear dynamic analysis methods for complex structures. Each method has advantages and shortcomings. Method of dynamic analysis for complex structure is selected considering characteristics of dynamic loading, computer facility available number of degree of freedem and accuracy of results. It is a main point of view to get economical results rather then accurate ones for analysis of general complex structures, Mode superposition method and direct integration method are generally used. However, the characteristics of load is not considered in mode superpositon method, the personal computer cannot be used in direct integration methods. To over-come these shortcomings, the component mode method incorporating Ritz algorithm updated is proposed to solve economically dynamic behavior of the structures. The purpose of study is a formulation of algorithm, and computer programing suitable for dynamic analysis of the complex structure in personal computer environment.

  • PDF

Calculation of Dynamic Stress Time History of a Component Using Computer Simulation (컴퓨터 시뮬레이션을 이용한 동응력 이력 계산기술 개발)

  • 박찬종;박태원
    • Journal of the Korean Society for Precision Engineering
    • /
    • v.17 no.1
    • /
    • pp.52-60
    • /
    • 2000
  • In order to design a reliable machine component efficiently, it is necessary to set up the process of durability analysis using computer simulation technique. In this paper, two methods for dynamic stress calculation, which are basis of durability analysis, are reviewed. Then, a user-oriented dynamic stress analysis program is developed from these two algorithms together with a general-purpose flexible body dynamic analysis and structural analysis programs. Finally, a slider-crank mechanism which has a flexible connecting-rod is chosen to show the special characteristics of these two dynamic stress calculation methods.

  • PDF

Design of Automobile Exhaust System using a Top-Down Approach Design Methodology (상하향식 설계법을 이용한 자동차 배기시스템의 설계)

  • 고병갑;박경진
    • Transactions of the Korean Society of Automotive Engineers
    • /
    • v.5 no.6
    • /
    • pp.13-27
    • /
    • 1997
  • In the modern design technology, a component should be designed to fit into the overall system performance. A design methodology is developed to expedite the mechan- ical design of complex mechanical systems, The relation between the system design and component design is defined by a top-down approach and the results from the system design are utilized in the component design process. As a design example, an automobile exhaust system is selected for the system design and a bellows is chosen for a component design. Design methodology based on the top-down approach consists of five steps; (1) Analysis of service load, (2) Development of a lumped parameter, (3) Completion of the system design, (4) Selection of the component topology, (5) Completion of the component design, A method using a equivalent matrix is developed in order to determine unknown external forces in linear structural analyses. The bellows is also analyzed by the finite element method using a conical frustum shell element. Various experiments are performed to verify the developed theories. The top-down desi- gn approach is demonstrated by a design case using structural and shape optimization technology. Since the method is relatively simple and easy compared to other methods, it can be applied to the general design where system and component designs are involves simultaneously.

  • PDF

Investigation of physical characteristics for Al2O3:C dosimeter using LM-OSL

  • Kim, Myung-Jin;Lee, Young-Ju;Kim, Ki-Bum;Hong, Duk-Geun
    • Analytical Science and Technology
    • /
    • v.32 no.4
    • /
    • pp.131-138
    • /
    • 2019
  • This paper reports results on the physical properties of a powder type of $Al_2O_3:C$ commonly used as a luminescence dosimeter using the LM-OSL technique. On the analysis with the general order kinetics model, the LM-OSL signal measured appeared to be composed of three components (fast, medium, slow) showing the largest area in the medium component. The photoionization cross sections of three components were distributed between $10^{-19}{\sim}10^{-21}cm^2$. The values of the thermal assistance energy were evaluated the largest in slow component and the smallest in fast component, which indicates the electrons trapped in defects attributed to slow component should be the most sensitive to thermal vibration among three components. In illumination to blue light, the fast component showed a rapid linear decay and completely disappeared after light exposure time of about 5 s. The medium component decayed with two exponential elements but the slow component did not observed any noticeable change until light exposure time of 40 s. In a dose response study, all components exhibited a linear behaviour up to approximately 10 Gy.

Simulation of Groundwater Variation Characteristics of Hancheon Watershed in Jeju Island using Integrated Hydrologic Modeling (통합수문모형을 이용한 제주 한천유역의 지하수 변동 특성 모의)

  • Kim, Nam-Won;Na, Hanna;Chung, Il-Moon
    • Journal of Environmental Science International
    • /
    • v.22 no.5
    • /
    • pp.515-522
    • /
    • 2013
  • To investigate groundwater variation characteristics in the Hancheon watershed, Jeju Island, an integrated hydrologic component analysis was carried out. For this purpose, SWAT-MODFLOW which is an integrated surface-groundwater model was applied to the watershed for continuous watershed hydrologic analysis as well as groundwater modeling. First, ephemeral stream characteristics of Hancheon watershed can be clearly simulated which is unlikely to be shown by a general watershed hydrologic model. Second, the temporally varied groundwater recharge can be properly obtained from SWAT and then spatially distributed groundwater recharge can be made by MODFLOW. Finally, the groundwater level variation was simulated with distributed groundwater pumping data. Since accurate recharge as well as abstraction can be reflected into the groundwater modeling, more realistic hydrologic component analysis and groundwater modeling could be possible.