This paper presents an improved Voice Activity Detection （VAD） algorithm which uses the Signal-to-Noise Ratio （SNR） measure. We assume that noise Power Spectral Density （PSD） in each spectral bin follows a Rayle...This paper presents an improved Voice Activity Detection （VAD） algorithm which uses the Signal-to-Noise Ratio （SNR） measure. We assume that noise Power Spectral Density （PSD） in each spectral bin follows a Rayleigh distribution. Rayleigh distributions with its asymmetric tail characteristics give a better description of the noise PSD distribution than Gaussian distribution. Under this asstlmption, a new threshold updating expression is derived. Since the analytical integral of the false alarm probability, the threshold updating expression can be represented without the inverse complementary error function and low computational complexity is achieved in our system. Experimental results show that the proposed VAD outperforms or at least is comparable with the VAD scheme presented by Davis under several noise environments and has a lower computational complexity.展开更多
Using observations from the Michelson Doppler Imager（MDI） onboard Solar and Heliospheric Observatory（SOHO）, we develop a computational algorithm to automatically identify bipolar magnetic regions（BMRs） in active...Using observations from the Michelson Doppler Imager（MDI） onboard Solar and Heliospheric Observatory（SOHO）, we develop a computational algorithm to automatically identify bipolar magnetic regions（BMRs） in active regions（ARs）, and then study their statistical properties.The individual magnetic（positive or negative） pole of a BMR is determined from the region with an absolute strength above 55 G and with an area larger than 250 pixel^2（~495 Mm^2）, while a BMR is identified as a pair of positive and negative poles with the shortest area-weight distance between them.Based on this method, 2234 BMRs are identified from MDI synoptic magnetograms between Carrington Rotations 1909（1996 May 06） and 2104（2010 December 10）. 1005 of them are located in the northern hemisphere, while the other 1229 are in the southern hemisphere. We find that the BMR parameters（e.g., latitude, separation, fragment number and strength） are similar to those of ARs. Moreover, based on the maximum likelihood estimation（MLE） method, the frequency distributions representing the occurrence of these BMRs as functions of area and magnetic flux exhibit a power-law behavior, i.e.,dN/dx ∝ x^（-αx）, with indices of α_A = 1.98 ± 0.06 and α_F = 1.93 ± 0.05 respectively. We also find that their orientation angles（θ） follow ＂Hale＇s Polarity Law＂ and deviate slightly toward the direction of the solar equator. Consistent with previous findings, we obtain the dependence of orientation angles on latitudes for normal BMRs during the 23 rd solar cycle. The north-south asymmetry of these BMRs is also detected here.展开更多
Solar flares strongly influence space weather and human activities, and their prediction is highly complex. The existing solutions such as data based approaches and model based approaches have a common shortcoming whi...Solar flares strongly influence space weather and human activities, and their prediction is highly complex. The existing solutions such as data based approaches and model based approaches have a common shortcoming which is the lack of human engagement in the forecasting process. An image-case-based reasoning method is introduced to achieve this goal. The image case library is com- posed of SOHO/MDI longitudinal magnetograms, the images from which exhibit the maximum hori- zontal gradient, the length of the neutral line and the number of singular points that are extracted for retrieving similar image cases. Genetic optimization algorithms are employed for optimizing the weight assignment for image features and the number of similar image cases retrieved. Similar image cases and prediction results derived by majority voting for these similar image cases are output and shown to the forecaster in order to integrate his/her experience with the final prediction results. Experimental results demonstrate that the case-based reasoning approach has slightly better performance than other methods, and is more efficient with forecasts improved by humans.展开更多
A periodicity analysis of the radio light curves of the blazar NRAO 530 at 14.5, 8.0, and 4.8 GHz is presented employing an improved phase dispersion mini- mization technique. The result, which shows two persistent pe...A periodicity analysis of the radio light curves of the blazar NRAO 530 at 14.5, 8.0, and 4.8 GHz is presented employing an improved phase dispersion mini- mization technique. The result, which shows two persistent periodic components of 6 and ~ 10 yr at all three frequencies, is consistent with the results obtained with the Lomb-Scargle periodogram and weighted wavelet Z-transform algorithms. The reliability of the derived periodicities is confirmed by the Monte Carlo numerical sim- ulations which show a high statistical confidence. （Quasi-）Periodic fluctuations of the radio luminosity of NRAO 530 might be associated with the oscillations of the accre- tion disk triggered by hydrodynamic instabilities of the accreted flow.展开更多
A texture image segmentation based on nonlinear diffusion is presented. The scale of texture can be measured during the process of nonlinear diffusion. A smooth 5-channel vector image with edge preserved, which is com...A texture image segmentation based on nonlinear diffusion is presented. The scale of texture can be measured during the process of nonlinear diffusion. A smooth 5-channel vector image with edge preserved, which is composed of intensity, scale and orientation of texture image, can be achieved by coupled nonlinear diffusion. A multi-channel statistical region active contour is employed to segment this vector image. The method can be seen as a kind of unsupervised segmentation because parameters are not sensitive to different texture images. Experimental results show its high efficiency in the semiautomatic extraction of texture image.展开更多
We made model fitting to the mid-to-far infrared spectral energy distributions （SEDs） for different categories of galaxies in the main extragalactic field of the Spitzer First Look Survey with the aid of spectroscop...We made model fitting to the mid-to-far infrared spectral energy distributions （SEDs） for different categories of galaxies in the main extragalactic field of the Spitzer First Look Survey with the aid of spectroscopic information from the Sloan Digital Sky Survey. We find that the mid-to-far infrared SEDs of HII galaxies, mixture type galaxies and LINERs can be well fitted by the one-parameter （α） dust model of Dale et al. plus the 13 Gyr dust-free elliptical galaxy model. The statistics of α values indicates that all these galaxies tend to be quiescent, although the HII galaxies are relatively more active than the LINERs. The midinfrared SEDs of absorption galaxies are well fitted simply by the 13 Gyr dust-free elliptical galaxy template, and the near-to-mid infrared SEDs of QSOs can be represented by AGN NGC 5506.展开更多
基金Supported by the National Natural Science Foundation of China (No. 60874060)
文摘This paper presents an improved Voice Activity Detection （VAD） algorithm which uses the Signal-to-Noise Ratio （SNR） measure. We assume that noise Power Spectral Density （PSD） in each spectral bin follows a Rayleigh distribution. Rayleigh distributions with its asymmetric tail characteristics give a better description of the noise PSD distribution than Gaussian distribution. Under this asstlmption, a new threshold updating expression is derived. Since the analytical integral of the false alarm probability, the threshold updating expression can be represented without the inverse complementary error function and low computational complexity is achieved in our system. Experimental results show that the proposed VAD outperforms or at least is comparable with the VAD scheme presented by Davis under several noise environments and has a lower computational complexity.
基金supported by the National Natural Science Foundation of China (Grant Nos. 11603077, 11573072 and 11333009)the Youth Fund of Jiangsu (No. BK20161095)the Laboratory (No. 2010DP173032)
文摘Using observations from the Michelson Doppler Imager（MDI） onboard Solar and Heliospheric Observatory（SOHO）, we develop a computational algorithm to automatically identify bipolar magnetic regions（BMRs） in active regions（ARs）, and then study their statistical properties.The individual magnetic（positive or negative） pole of a BMR is determined from the region with an absolute strength above 55 G and with an area larger than 250 pixel^2（~495 Mm^2）, while a BMR is identified as a pair of positive and negative poles with the shortest area-weight distance between them.Based on this method, 2234 BMRs are identified from MDI synoptic magnetograms between Carrington Rotations 1909（1996 May 06） and 2104（2010 December 10）. 1005 of them are located in the northern hemisphere, while the other 1229 are in the southern hemisphere. We find that the BMR parameters（e.g., latitude, separation, fragment number and strength） are similar to those of ARs. Moreover, based on the maximum likelihood estimation（MLE） method, the frequency distributions representing the occurrence of these BMRs as functions of area and magnetic flux exhibit a power-law behavior, i.e.,dN/dx ∝ x^（-αx）, with indices of α_A = 1.98 ± 0.06 and α_F = 1.93 ± 0.05 respectively. We also find that their orientation angles（θ） follow ＂Hale＇s Polarity Law＂ and deviate slightly toward the direction of the solar equator. Consistent with previous findings, we obtain the dependence of orientation angles on latitudes for normal BMRs during the 23 rd solar cycle. The north-south asymmetry of these BMRs is also detected here.
基金supported by the National Natural Science Foundation of China (Grant No. 11078010)
文摘Solar flares strongly influence space weather and human activities, and their prediction is highly complex. The existing solutions such as data based approaches and model based approaches have a common shortcoming which is the lack of human engagement in the forecasting process. An image-case-based reasoning method is introduced to achieve this goal. The image case library is com- posed of SOHO/MDI longitudinal magnetograms, the images from which exhibit the maximum hori- zontal gradient, the length of the neutral line and the number of singular points that are extracted for retrieving similar image cases. Genetic optimization algorithms are employed for optimizing the weight assignment for image features and the number of similar image cases retrieved. Similar image cases and prediction results derived by majority voting for these similar image cases are output and shown to the forecaster in order to integrate his/her experience with the final prediction results. Experimental results demonstrate that the case-based reasoning approach has slightly better performance than other methods, and is more efficient with forecasts improved by humans.
基金supported by the National Basic Research Programof China (973 Program,2009CB24900)the Guangxi Natural Science Foundation (0991018Z)+3 种基金the Science and Technology Commission of Shanghai Municipality (06DZ22101)supported by a grant from the Chinese Academy of Sciencesa visitor grant from the The Netherlands Science Foundationsupported by funds from the NSF,NASA,and the University of Michigan
文摘A periodicity analysis of the radio light curves of the blazar NRAO 530 at 14.5, 8.0, and 4.8 GHz is presented employing an improved phase dispersion mini- mization technique. The result, which shows two persistent periodic components of 6 and ~ 10 yr at all three frequencies, is consistent with the results obtained with the Lomb-Scargle periodogram and weighted wavelet Z-transform algorithms. The reliability of the derived periodicities is confirmed by the Monte Carlo numerical sim- ulations which show a high statistical confidence. （Quasi-）Periodic fluctuations of the radio luminosity of NRAO 530 might be associated with the oscillations of the accre- tion disk triggered by hydrodynamic instabilities of the accreted flow.
文摘A texture image segmentation based on nonlinear diffusion is presented. The scale of texture can be measured during the process of nonlinear diffusion. A smooth 5-channel vector image with edge preserved, which is composed of intensity, scale and orientation of texture image, can be achieved by coupled nonlinear diffusion. A multi-channel statistical region active contour is employed to segment this vector image. The method can be seen as a kind of unsupervised segmentation because parameters are not sensitive to different texture images. Experimental results show its high efficiency in the semiautomatic extraction of texture image.
基金Supported by the National Natural Science Foundation of China.
文摘We made model fitting to the mid-to-far infrared spectral energy distributions （SEDs） for different categories of galaxies in the main extragalactic field of the Spitzer First Look Survey with the aid of spectroscopic information from the Sloan Digital Sky Survey. We find that the mid-to-far infrared SEDs of HII galaxies, mixture type galaxies and LINERs can be well fitted by the one-parameter （α） dust model of Dale et al. plus the 13 Gyr dust-free elliptical galaxy model. The statistics of α values indicates that all these galaxies tend to be quiescent, although the HII galaxies are relatively more active than the LINERs. The midinfrared SEDs of absorption galaxies are well fitted simply by the 13 Gyr dust-free elliptical galaxy template, and the near-to-mid infrared SEDs of QSOs can be represented by AGN NGC 5506.
