• 제목/요약/키워드: spectral analysis.

검색결과 2,986건 처리시간 0.034초

Considerations on the Artifacts in a Cerebral Vascular Audiospectroscopy (CeVASS) using Spectral and Cross-Spectral Analysis

  • Kato, Yuki;Hasegawa, Hiroshi;Kasuga, Masao;Kyoshima, Kazuhiko;Muraoka, Shinsuke
    • 대한전자공학회:학술대회논문집
    • /
    • 대한전자공학회 2000년도 ITC-CSCC -1
    • /
    • pp.221-224
    • /
    • 2000
  • The purposes of this paper are to clarify the influences of artifacts on a cerebrovascular audiospectroscopy (CeVASS) and to improve the detecting precision of the CeVASS. First, we perform the CeVASS in 15 healthy adults with various measuring positions such as forehead eyes, cervical parts, temple parts, and occipitomastoid parts, and the influences of artifacts on each position is estimated. Next, we discuss about the removal of the artifacts using cross-spectral analysis. Finally, we propose an improved system of the CeVASS.

  • PDF

Principal Componet Analysis에 의한 고 분해능 AR 모델링과 스텍트럼 추정 (High Resolution AR Spectral Estimation by Principal Component Analysis)

  • 양흥석;이석원;공성곤
    • 대한전기학회논문지
    • /
    • 제36권11호
    • /
    • pp.813-818
    • /
    • 1987
  • 본 논문에서는 AR모델링 기법과 Principal Component Analysis를 이용하여 주파수 분해능이 좋은 스펙트럼 추정방법을 제안하였다. 주어진 데이타를 고유벡터에 의해 전개하고, 각 고유벡터에 대하여 AR스펙트럼추정기법에 의하여 고유 스펙트럼을 구하고, 이것을 고유치로 가중하여 합성함으로써 주파수 분해능력이 좋은 스펙트럼 추정치를 얻을 수 있다. 계산량은 증가하지만, 특히 짧은 데이타, 협대역 신호, 그리고 신호대 잡음비가 낮은 경우에 비헤서도 정확한 스펙트럼 추정이 가능하다는 것을 컴퓨터 시뮬레이션을 통해 확인 하였다.

이산 웨이브릿 변환을 이용한 소나 자료처리에 관한 연구 (A Study on the Sonar Data Processing by Using a Discrete Wavelet Transform)

  • 김진후;김현도
    • 한국해양공학회:학술대회논문집
    • /
    • 한국해양공학회 2003년도 춘계학술대회 논문집
    • /
    • pp.324-329
    • /
    • 2003
  • Spectral analysis is an important signal processing tool for time series data. The transformation of a time series into the frequency domain is the basis for a significant number of processing algorithms and interpretive methods. Recently developed transforms based on the new mathematical field of wavelet analysis bypass the resolution limitation and offer superior spectral decomposition. The discrete wavelet transform of Sonar data provides spectral localization of noises, hence noises can be filtered out successfully.

  • PDF

가변 윈도우를 이용한 CELP 부호화기의 음질 향상에 관한 연구 (A Study on a Improvement of the Speech Quality with Variable Window in CELP Vocoder)

  • 주상규
    • 한국산학기술학회:학술대회논문집
    • /
    • 한국산학기술학회 2010년도 춘계학술발표논문집 1부
    • /
    • pp.265-268
    • /
    • 2010
  • There have been proposed two types of low bit rate vocoder upto now : One is MBE type using the spectrum modeling and another is CELP type using the hybrid coding method. CELP type vocoder has mainly studied between them. Specially, much of intensity is concentrated in CELP vocoder due to the emergence of Internet Phone and PCS in a domestic. In order to improve the speech quality in CELP vocoder, in this paper, we proposed a new spectrum analysis algorithm with variable window. In CELP vocoder, the spectrum of the synthesised speech signal is distorted because the fixed size windows is used for spectrum analysis. So we have measured the spectral leakage and in order to minimize the spectral leakage have adjusted the window size. Applying this method G.723.1 ACELP, we can get SD(Spectral Distortion) reduction 0.084(dB), residual energy reduction 6.3% and MOS(Mean Opinion Score) improvement 0.1.

  • PDF

Application of the exact spectral element method in the analysis of the smart functionally graded plate

  • Farhad Abad;Jafar Rouzegar;Saeid Lotfian
    • Steel and Composite Structures
    • /
    • 제47권2호
    • /
    • pp.297-313
    • /
    • 2023
  • This study aims to extend the application of the spectral element method (SEM) to wave propagation and free vibration analysis of functionally graded (FG) plates integrated with thin piezoelectric layers, plates with tapered thickness and structure on elastic foundations. Also, the dynamic response of the smart FG plate under impact and moving loads is presented. In this paper, the dynamic stiffness matrix of the smart rectangular FG plate is determined by using the exact dynamic shape functions based on Mindlin plate assumptions. The low computational time and results' independence with the number of elements are two significant features of the SEM. Also, to prove the accuracy and efficiency of the SEM, results are compared with Abaqus simulations and those reported in references. Furthermore, the effects of boundary conditions, power-law index, piezoelectric layers thickness, and type of loading on the results are studied.

기적신호의 스펙트럼 분석을 통한 음향수신장치의 방향탐지 알고리즘 설계 및 분석 (Design and analysis of direction indicating algorithm for sound reception system based on spectral analysis of whistle signal)

  • 권혁진;김정창
    • Journal of Advanced Marine Engineering and Technology
    • /
    • 제41권1호
    • /
    • pp.83-90
    • /
    • 2017
  • 본 논문에서는 스펙트럼 분석을 통해 선박 기적신호의 위상 차이를 이용한 음향수신장치의 방향탐지 알고리즘을 설계하고 분석한다. 제안하는 시스템은 한 지점에서 90도 간격의 네 방향을 향하도록 설치된 마이크로폰으로부터 기적신호를 수신한다. 제안하는 알고리즘은 스펙트럼 분석을 통하여 각 수신 신호의 위상을 검출하고 인접한 2개의 마이크로폰으로부터 수신된 신호의 위상 차이를 얻음으로써 기적신호의 방향을 추정한다. 또한, 수신 신호의 도착 각도에 따라 인접한 두 수신 신호 사이의 위상 차이를 이론적으로 분석하고 DSP 칩을 이용하여 제안하는 시스템을 구현한다. 뿐만 아니라, 구현된 시스템을 이용하여 실험실 환경에서 제안하는 알고리즘의 동작을 검증한다. 실험 결과는 제안하는 방식이 기적신호의 방향을 잘 추정할 수 있음을 보여준다.

Spectral Sensing for Plant Stress Assessment - A Review -

  • Kim, Y.;Reid, J.F.
    • Agricultural and Biosystems Engineering
    • /
    • 제7권1호
    • /
    • pp.27-41
    • /
    • 2006
  • Assessment of nitrogen and chlorophyll content from crop leaves can help growers adjust N fertilizer rates to meet the demands of the crop. Numerous researchers have presented their studies about spectral signature of plant leaves to characterize the plant features. However, interrelational review and summary were limited and a communication gap exists between the plant science and optical engineering. Understanding the mechanism of leaf interaction to electromagnetic radiation and factors affecting spectrophotometric measurements can enhance the foundation of optical remote sensing technologies. This paper provides extensive review of previous works in optical sensing and explains the basics of plant optics, spectral measurements for plant stress, factors that affect sensitivity to spectral analysis, and applications that deploy optical remote sensing technologies.

  • PDF

헤드램프의 시계성 평가를 위한 야간 도로 영상 재현 알고리즘 (An reproduction algorithm of nighttime road-image for visibility evaluation of headlamps)

  • 이철희;하영호
    • 대한전자공학회:학술대회논문집
    • /
    • 대한전자공학회 2000년도 추계종합학술대회 논문집(4)
    • /
    • pp.69-72
    • /
    • 2000
  • This study proposes a new calculation method for generating real nighttime lamp-lit images. In order to improve the color appearance in the prediction of a nighttime lamp-lighted scene, the lamp-lit image is synthesized based on spectral distribution using the estimated local spectral distribution of the headlamps and the surface reflectance of every object. The principal component analysis method is introduced to estimate the surface color of an object, and the local spectral distribution of the headlamps is calculated based on the illuminance data and spectral distribution of the illuminating headlamps. HID and halogen lamps are utilized to create beam patterns and captured road scenes are used as background images to simulate actual headlamp-lit images on a monitor. As a result, the reproduced images presented a color appearance that was very close to a real nighttime road image illuminated by single and multiple headlamps.

  • PDF

Hyperion 영상의 제약선형분광혼합분석 기반 무감독 Endmember 추출 최적화 기법 (Unsupervised Endmember Selection Optimization Process based on Constrained Linear Spectral Unmixing of Hyperion Image)

  • 최재완;김용일;유기윤
    • 한국측량학회:학술대회논문집
    • /
    • 한국측량학회 2006년도 춘계학술발표회 논문집
    • /
    • pp.211-216
    • /
    • 2006
  • The Constrained Linear Spectral Unmixing(CLSU) is investigated for sub-pixel image processing, Its result is the abundance map which mean fractions of endmember existing in a mixed pixel. Compared to the Linear Spectral Unmixing using least square method, CLSU uses the NNLS (Non-Negative Least Square) algorithm to guarantee that the estimated fractions are constrained. But, CLSU gets Into difficulty in image processing due to select endmember at a user's disposition. In this study, endmember selection optimization method using entropy in the error-image analysis is proposed. In experiments which is used hyperion image, it is shown that our method can select endmember number than CLSU based on unsupervised endemeber selection.

  • PDF

비정질 실리콘 박막 트랜지스터의 광특성 분석을 위한 백라이트의 광자 에너지 스펙트럼에 대한 연구 (A Study on the Photon Energy Spectrums of Backlight for the Analysis of the Photoelectric Characteristics of a-Si:H TFT)

  • 정경서;권상직;조의식
    • 한국전기전자재료학회논문지
    • /
    • 제22권12호
    • /
    • pp.1058-1062
    • /
    • 2009
  • For the investigation of the mechanism of photoelectric characteristics of a hydrogenated amorphous silicon thin film transistor(a-Si:H TFT), spectral characteristics of various backlights were analyzed in terms of the photon energy at each wavelength. Photon energy spectral characteristics were obtained through the multiplication of each photon energy and spectral intensities of backlights at each wavelength and the total photon energies were obtained by the integration of the photon energy spectrums. From the comparison of the experimental photo leakage current and the calculated photon energy, it was possible to conclude that the absorption of illuminated backlight to a-Si:H layer and the generation of electrons and holes are mainly carried out at the wavelength less than 500 nm as described in previous reports.