Gaze estimation,a crucial non-verbal communication cue,has achieved remarkable progress through convolutional neural networks.However,accurate gaze prediction in uncon-strained environments,particularly in extreme hea...Gaze estimation,a crucial non-verbal communication cue,has achieved remarkable progress through convolutional neural networks.However,accurate gaze prediction in uncon-strained environments,particularly in extreme head poses,partial occlusions,and abnormal lighting,remains challenging.Existing models often struggle to effectively focus on discriminative ocular features,leading to suboptimal performance.To address these limitations,this paper proposes dual-branch gaze estimation with Gaussian mixture distribution heatmaps and dynamic adaptive loss function(DMGDL),a novel dual-branch gaze estimation algorithm.By introducing Gaussian mixture distribution heatmaps centered on pupil positions as spatial attention guides,the model is enabled to prioritize ocular regions.Additionally,a dual-branch network architecture is designed to separately extract features for yaw and pitch angles,enhancing flexibility and mitigating cross-angle interference.A dynamic adaptive loss function is further formulated to address discontinuities in angle estimation,improving robustness and convergence stability.Experimental evaluations on three benchmark datasets demonstrate that DMGDL outperforms state-of-the-art methods,achiev-ing a mean angular error of 3.98°on the Max-Planck institute for informatics face gaze(MPI-IFaceGaze)dataset,10.21°on the physically unconstrained gaze estimation in the wild(Gaze360)dataset and 6.14°on the real-time eye gaze estimation in natural environments(RT-Gene)dataset,exhibiting superior generalization and robustness.展开更多
The computational accuracy and efficiency of modeling the stress spectrum derived from bridge monitoring data significantly influence the fatigue life assessment of steel bridges.Therefore,determining the optimal stre...The computational accuracy and efficiency of modeling the stress spectrum derived from bridge monitoring data significantly influence the fatigue life assessment of steel bridges.Therefore,determining the optimal stress spectrum model is crucial for further fatigue reliability analysis.This study investigates the performance of the REBMIX algorithm in modeling both univariate(stress range)and multivariate(stress range and mean stress)distributions of the rain-flowmatrix for a steel arch bridge,usingAkaike’s Information Criterion(AIC)as a performance metric.Four types of finitemixture distributions—Normal,Lognormal,Weibull,and Gamma—are employed tomodel the stress range.Additionally,mixed distributions,including Normal-Normal,Lognormal-Normal,Weibull-Normal,and Gamma-Normal,are utilized to model the joint distribution of stress range and mean stress.The REBMIX algorithm estimates the number of components,component weights,and component parameters for each candidate finite mixture distribution.The results demonstrate that the REBMIX algorithm-based mixture parameter estimation approach effectively identifies the optimal distribution based on AIC values.Furthermore,the algorithm exhibits superior computational efficiency compared to traditional methods,making it highly suitable for practical applications.展开更多
Aiming at the problem of filtering precision degradation caused by the random outliers of process noise and measurement noise in multi-target tracking(MTT) system,a new Gaussian-Student’s t mixture distribution proba...Aiming at the problem of filtering precision degradation caused by the random outliers of process noise and measurement noise in multi-target tracking(MTT) system,a new Gaussian-Student’s t mixture distribution probability hypothesis density(PHD) robust filtering algorithm based on variational Bayesian inference(GST-vbPHD) is proposed.Firstly,since it can accurately describe the heavy-tailed characteristics of noise with outliers,Gaussian-Student’s t mixture distribution is employed to model process noise and measurement noise respectively.Then Bernoulli random variable is introduced to correct the likelihood distribution of the mixture probability,leading hierarchical Gaussian distribution constructed by the Gaussian-Student’s t mixture distribution suitable to model non-stationary noise.Finally,the approximate solutions including target weights,measurement noise covariance and state estimation error covariance are obtained according to variational Bayesian inference approach.The simulation results show that,in the heavy-tailed noise environment,the proposed algorithm leads to strong improvements over the traditional PHD filter and the Student’s t distribution PHD filter.展开更多
In this paper,a finite mixture of m-Brlang distributions is proposed.Different moments,shape characteristics and parameter estimates of the proposed model are also provided.The proposed mixture has the property that i...In this paper,a finite mixture of m-Brlang distributions is proposed.Different moments,shape characteristics and parameter estimates of the proposed model are also provided.The proposed mixture has the property that it has a bounded hazard function.A special case of the mixed Erlang distribution is introduced and discussed.In addition,a predictive technique is introduced to estimate the needed number of mixture components to fit a certain data.A real data concerning the confirmed COVID-19 cases in Egypt is introduced to utilize the predictive estimation technique.Two more real datasets are used to examine the flexibility of the proposed model.展开更多
In this paper, we propose a robust mixture regression model based on the skew scale mixtures of normal distributions (RMR-SSMN) which can accommodate asymmetric, heavy-tailed and contaminated data better. For the vari...In this paper, we propose a robust mixture regression model based on the skew scale mixtures of normal distributions (RMR-SSMN) which can accommodate asymmetric, heavy-tailed and contaminated data better. For the variable selection problem, the penalized likelihood approach with a new combined penalty function which balances the SCAD and l<sub>2</sub> penalty is proposed. The adjusted EM algorithm is presented to get parameter estimates of RMR-SSMN models at a faster convergence rate. As simulations show, our mixture models are more robust than general FMR models and the new combined penalty function outperforms SCAD for variable selection. Finally, the proposed methodology and algorithm are applied to a real data set and achieve reasonable results.展开更多
In this paper, we provide a method based on quantiles to estimate the parameters of a finite mixture of Fréchet distributions, for a large sample of strongly dependent data. This is a situation that appears when ...In this paper, we provide a method based on quantiles to estimate the parameters of a finite mixture of Fréchet distributions, for a large sample of strongly dependent data. This is a situation that appears when dealing with environmental data and there was a real need of such method. We validate our approach by means of estimation and goodness-of-fit testing over simulated data, showing an accurate performance.展开更多
Highly versatile machines, such as wheel loaders, forklifts, and mining haulers, are subject to many kinds of working conditions, as well as indefinite factors that lead to the complexity of the load. The load probabi...Highly versatile machines, such as wheel loaders, forklifts, and mining haulers, are subject to many kinds of working conditions, as well as indefinite factors that lead to the complexity of the load. The load probability distribution function （PDF） of transmission gears has many distributions centers; thus, its PDF cannot be well represented by just a single-peak function. For the purpose of representing the distribution characteristics of the complicated phenomenon accurately, this paper proposes a novel method to establish a mixture model. Based on linear regression models and correlation coefficients, the proposed method can be used to automatically select the best-fitting function in the mixture model. Coefficient of determination, the mean square error, and the maximum deviation are chosen and then used as judging criteria to describe the fitting precision between the theoretical distribution and the corresponding histogram of the available load data. The applicability of this modeling method is illustrated by the field testing data of a wheel loader. Meanwhile, the load spectra based on the mixture model are compiled. The comparison results show that the mixture model is more suitable for the description of the load-distribution characteristics. The proposed research improves the flexibility and intelligence of modeling, reduces the statistical error and enhances the fitting accuracy, and the load spectra complied by this method can better reflect the actual load characteristic of the gear component.展开更多
In this paper, we introduce a modification of the Quasi Lindley distribution which has various advantageous properties for the lifetime data. Several fundamental structural properties of the distribution are explored....In this paper, we introduce a modification of the Quasi Lindley distribution which has various advantageous properties for the lifetime data. Several fundamental structural properties of the distribution are explored. Its density function can be left-skewed, symmetrical, and right-skewed shapes with various rages of tail-weights and dispersions. The failure rate function of the new dist</span><span style="font-family:Verdana;">ribution has the flexibility to be increasing, decreasing, constant, an</span><span style="font-family:Verdana;">d bathtub shapes. A simulation study is done to examine the performance of maximum likelihood and moment estimation methods in its unknown parameter estimations based on the asymptotic theory. The potentiality of the new distribution is illustrated by means of applications to the simulated and three real-world data sets.展开更多
The fluctuation characteristics is the inherent property of wind power.Through analysis of a large number of wind t＇anns based on measured data,we find it describes the best probability distribution of wind power flu...The fluctuation characteristics is the inherent property of wind power.Through analysis of a large number of wind t＇anns based on measured data,we find it describes the best probability distribution of wind power fluctuation for the mixed Gauss distribution of two components,and try to carry out the physical interpretation of two components.Further discussion is between the probability distribution of fluctuating wind power time difference and whole relationship.It is found that the two have basic similarity.Through comparing the different time level data quantified losses the information of wind power fluctuation,quantitative determination of the degree of impact prediction.We can summarize and understand of wind power fluctuation,constructing instance from the wind farm construction and monitoring prediction two aspect recommendations to overcome the adverse effects of wind power fluctuations on the power grid operation.展开更多
In Wyner-Ziv (WZ) Distributed Video Coding (DVC), correlation noise model is often used to describe the error distribution between WZ frame and the side information. The accuracy of the model can influence the perform...In Wyner-Ziv (WZ) Distributed Video Coding (DVC), correlation noise model is often used to describe the error distribution between WZ frame and the side information. The accuracy of the model can influence the performance of the video coder directly. A mixture correlation noise model in Discrete Cosine Transform (DCT) domain for WZ video coding is established in this paper. Different correlation noise estimation method is used for direct current and alternating current coefficients. Parameter estimation method based on expectation maximization algorithm is used to estimate the Laplace distribution center of direct current frequency band and Mixture Laplace-Uniform Distribution Model (MLUDM) is established for alternating current coefficients. Experimental results suggest that the proposed mixture correlation noise model can describe the heavy tail and sudden change of the noise accurately at high rate and make significant improvement on the coding efficiency compared with the noise model presented by DIStributed COding for Video sERvices (DISCOVER).展开更多
In order to analyze the microstructure of salt anti-freezing asphalt concrete, i e, MFL（Mafilon） modified asphalt concrete, MIP（mercury intrusion porosity） method was used to obtain the data including porosity and...In order to analyze the microstructure of salt anti-freezing asphalt concrete, i e, MFL（Mafilon） modified asphalt concrete, MIP（mercury intrusion porosity） method was used to obtain the data including porosity and pore size distribution in micro scale. Results show that the porosity grows up with the increase of immersion duration and the salt content. During the immersion, the amount of large pores（60-200 μm） grow up gradually and porosity also grows up correspondingly. Even with different immersion duration, most pores＇ size distribute is beyond 7000 nm.展开更多
A new family of univariate exponential slash distribution is introduced, which is based on elliptical distributions and defined by means of a stochastic representation as the scale mixture of an elliptically distribut...A new family of univariate exponential slash distribution is introduced, which is based on elliptical distributions and defined by means of a stochastic representation as the scale mixture of an elliptically distributed random variable with respect to the power of an exponential random variable. The same idea is extended to the multivariate case. General properties of the resulting families, including their moments and kurtosis coefficient, are stud- ied. And inferences based on methods of moment and maximum likelihood are discussed. A real data is presented to show this family is flexible and fits much better than other related families.展开更多
The classical risk process that is perturbed by diffusion is studied. The explicit expressions for the ruin probability and the surplus distribution of the risk process at the time of ruin are obtained when the claim ...The classical risk process that is perturbed by diffusion is studied. The explicit expressions for the ruin probability and the surplus distribution of the risk process at the time of ruin are obtained when the claim amount distribution is a finite mixture of exponential distributions or a Gamma (2, α) distribution.展开更多
This paper compares the stock return distribution models of mixture normal distribution, mixed diffusion-jump and GARCH models based on the data of Chinese stock market. The Schwarz criterion is also used. We find all...This paper compares the stock return distribution models of mixture normal distribution, mixed diffusion-jump and GARCH models based on the data of Chinese stock market. The Schwarz criterion is also used. We find all these models can capture the features of stock returns partly. EGARCH model is the best fitting to daily return and stable during different period. When the weekly and monthly returns are tested, the differences of the models＇ fitness become unobvious and unstable.展开更多
Mixture of Experts(MoE)regression models are widely studied in statistics and machine learning for modeling heterogeneity in data for regression,clustering and classification.Laplace distribution is one of the most im...Mixture of Experts(MoE)regression models are widely studied in statistics and machine learning for modeling heterogeneity in data for regression,clustering and classification.Laplace distribution is one of the most important statistical tools to analyze thick and tail data.Laplace Mixture of Linear Experts(LMoLE)regression models are based on the Laplace distribution which is more robust.Similar to modelling variance parameter in a homogeneous population,we propose and study a new novel class of models:heteroscedastic Laplace mixture of experts regression models to analyze the heteroscedastic data coming from a heterogeneous population in this paper.The issues of maximum likelihood estimation are addressed.In particular,Minorization-Maximization(MM)algorithm for estimating the regression parameters is developed.Properties of the estimators of the regression coefficients are evaluated through Monte Carlo simulations.Results from the analysis of two real data sets are presented.展开更多
Performance of Turbo-Codes in communication channels with impulsive noise is analyzed. First, mathematical model of impulsive noise is presented because it has non-Gaussian nature and is found in many wireless channel...Performance of Turbo-Codes in communication channels with impulsive noise is analyzed. First, mathematical model of impulsive noise is presented because it has non-Gaussian nature and is found in many wireless channels due to impulsive phenomena of radio-frequency interference. Then, with linear Log-MAP decoding algorithm for its low complexity, Turbo-Codes are adopted and analyzed in such communication channels. To confirm the performance of the proposed method, simulations on both static and fully interleaved flat Rayleigh fading channels with impulsive noise have been carried out. It is shown that Turbo-Codes have a better performance than the conventional methods （e.g. convolutionally coded system）.展开更多
The major male sterile genes in a new photo/thermo-sensitive genie male sterile (PTGMS) line B06S of rice were analyzed by the manipulation of mixture distribution theory. The results indicated that a pair of major ma...The major male sterile genes in a new photo/thermo-sensitive genie male sterile (PTGMS) line B06S of rice were analyzed by the manipulation of mixture distribution theory. The results indicated that a pair of major male sterile nuclear genes with large effects were responsible for controlling the male sterility of B06S.展开更多
The paper presents an algorithm of automatic target detection in Synthetic Aperture Radar(SAR) images based on Maximum A Posteriori(MAP). The algorithm is divided into three steps. First, it employs Gaussian mixture d...The paper presents an algorithm of automatic target detection in Synthetic Aperture Radar(SAR) images based on Maximum A Posteriori(MAP). The algorithm is divided into three steps. First, it employs Gaussian mixture distribution to approximate and estimate multi-modal histogram of SAR image. Then, based on the principle of MAP, when a priori probability is both unknown and learned respectively, the sample pixels are classified into different classes c = {target,shadow, background}. Last, it compares the results of two different target detections. Simulation results preferably indicate that the presented algorithm is fast and robust, with the learned a priori probability, an approach to target detection is reliable and promising.展开更多
Generalized method of moments based on probability generating function is considered. Estimation and model testing are unified using this approach which also leads to distribution free chi-square tests. The estimation...Generalized method of moments based on probability generating function is considered. Estimation and model testing are unified using this approach which also leads to distribution free chi-square tests. The estimation methods developed are also related to estimation methods based on generalized estimating equations but with the advantage of having statistics for model testing. The methods proposed overcome numerical problems often encountered when the probability mass functions have no closed forms which prevent the use of maximum likelihood (ML) procedures and in general, ML procedures do not lead to distribution free model testing statistics.展开更多
Hybrid clustering combines partitional and hierarchical clustering for computational effectiveness and versatility in cluster shape. In such clustering, a dissimilarity measure plays a crucial role in the hierarchical...Hybrid clustering combines partitional and hierarchical clustering for computational effectiveness and versatility in cluster shape. In such clustering, a dissimilarity measure plays a crucial role in the hierarchical merging. The dissimilarity measure has great impact on the final clustering, and data-independent properties are needed to choose the right dissimilarity measure for the problem at hand. Properties for distance-based dissimilarity measures have been studied for decades, but properties for density-based dissimilarity measures have so far received little attention. Here, we propose six data-independent properties to evaluate density-based dissimilarity measures associated with hybrid clustering, regarding equality, orthogonality, symmetry, outlier and noise observations, and light-tailed models for heavy-tailed clusters. The significance of the properties is investigated, and we study some well-known dissimilarity measures based on Shannon entropy, misclassification rate, Bhattacharyya distance and Kullback-Leibler divergence with respect to the proposed properties. As none of them satisfy all the proposed properties, we introduce a new dissimilarity measure based on the Kullback-Leibler information and show that it satisfies all proposed properties. The effect of the proposed properties is also illustrated on several real and simulated data sets.展开更多
基金supported by the Key Project of the NationalLanguage Commission(No.ZDI145-110)the AcademicResearch Projects of Beijing Union University(No.ZK20202514)+1 种基金the Key Laboratory Project(No.YYZN-2024-6)the Project for the Construction and Support of High-Level Innovative Teams in Beijing Municipal Institutions(No.BPHR20220121).
文摘Gaze estimation,a crucial non-verbal communication cue,has achieved remarkable progress through convolutional neural networks.However,accurate gaze prediction in uncon-strained environments,particularly in extreme head poses,partial occlusions,and abnormal lighting,remains challenging.Existing models often struggle to effectively focus on discriminative ocular features,leading to suboptimal performance.To address these limitations,this paper proposes dual-branch gaze estimation with Gaussian mixture distribution heatmaps and dynamic adaptive loss function(DMGDL),a novel dual-branch gaze estimation algorithm.By introducing Gaussian mixture distribution heatmaps centered on pupil positions as spatial attention guides,the model is enabled to prioritize ocular regions.Additionally,a dual-branch network architecture is designed to separately extract features for yaw and pitch angles,enhancing flexibility and mitigating cross-angle interference.A dynamic adaptive loss function is further formulated to address discontinuities in angle estimation,improving robustness and convergence stability.Experimental evaluations on three benchmark datasets demonstrate that DMGDL outperforms state-of-the-art methods,achiev-ing a mean angular error of 3.98°on the Max-Planck institute for informatics face gaze(MPI-IFaceGaze)dataset,10.21°on the physically unconstrained gaze estimation in the wild(Gaze360)dataset and 6.14°on the real-time eye gaze estimation in natural environments(RT-Gene)dataset,exhibiting superior generalization and robustness.
基金jointly supported by the Fundamental Research Funds for the Central Universities(Grant No.xzy012023075)the Zhejiang Engineering Research Center of Intelligent Urban Infrastructure(Grant No.IUI2023-YB-12).
文摘The computational accuracy and efficiency of modeling the stress spectrum derived from bridge monitoring data significantly influence the fatigue life assessment of steel bridges.Therefore,determining the optimal stress spectrum model is crucial for further fatigue reliability analysis.This study investigates the performance of the REBMIX algorithm in modeling both univariate(stress range)and multivariate(stress range and mean stress)distributions of the rain-flowmatrix for a steel arch bridge,usingAkaike’s Information Criterion(AIC)as a performance metric.Four types of finitemixture distributions—Normal,Lognormal,Weibull,and Gamma—are employed tomodel the stress range.Additionally,mixed distributions,including Normal-Normal,Lognormal-Normal,Weibull-Normal,and Gamma-Normal,are utilized to model the joint distribution of stress range and mean stress.The REBMIX algorithm estimates the number of components,component weights,and component parameters for each candidate finite mixture distribution.The results demonstrate that the REBMIX algorithm-based mixture parameter estimation approach effectively identifies the optimal distribution based on AIC values.Furthermore,the algorithm exhibits superior computational efficiency compared to traditional methods,making it highly suitable for practical applications.
基金Supported by the National Natural Science Foundation of China(No.61976080)the Science and Technology Key Project of Science and Technology Department of Henan Province(No.212102310298)the Innovation and Quality Improvement Project for Graduate Education of Henan University(No.SYL20010101)。
文摘Aiming at the problem of filtering precision degradation caused by the random outliers of process noise and measurement noise in multi-target tracking(MTT) system,a new Gaussian-Student’s t mixture distribution probability hypothesis density(PHD) robust filtering algorithm based on variational Bayesian inference(GST-vbPHD) is proposed.Firstly,since it can accurately describe the heavy-tailed characteristics of noise with outliers,Gaussian-Student’s t mixture distribution is employed to model process noise and measurement noise respectively.Then Bernoulli random variable is introduced to correct the likelihood distribution of the mixture probability,leading hierarchical Gaussian distribution constructed by the Gaussian-Student’s t mixture distribution suitable to model non-stationary noise.Finally,the approximate solutions including target weights,measurement noise covariance and state estimation error covariance are obtained according to variational Bayesian inference approach.The simulation results show that,in the heavy-tailed noise environment,the proposed algorithm leads to strong improvements over the traditional PHD filter and the Student’s t distribution PHD filter.
文摘In this paper,a finite mixture of m-Brlang distributions is proposed.Different moments,shape characteristics and parameter estimates of the proposed model are also provided.The proposed mixture has the property that it has a bounded hazard function.A special case of the mixed Erlang distribution is introduced and discussed.In addition,a predictive technique is introduced to estimate the needed number of mixture components to fit a certain data.A real data concerning the confirmed COVID-19 cases in Egypt is introduced to utilize the predictive estimation technique.Two more real datasets are used to examine the flexibility of the proposed model.
文摘In this paper, we propose a robust mixture regression model based on the skew scale mixtures of normal distributions (RMR-SSMN) which can accommodate asymmetric, heavy-tailed and contaminated data better. For the variable selection problem, the penalized likelihood approach with a new combined penalty function which balances the SCAD and l<sub>2</sub> penalty is proposed. The adjusted EM algorithm is presented to get parameter estimates of RMR-SSMN models at a faster convergence rate. As simulations show, our mixture models are more robust than general FMR models and the new combined penalty function outperforms SCAD for variable selection. Finally, the proposed methodology and algorithm are applied to a real data set and achieve reasonable results.
文摘In this paper, we provide a method based on quantiles to estimate the parameters of a finite mixture of Fréchet distributions, for a large sample of strongly dependent data. This is a situation that appears when dealing with environmental data and there was a real need of such method. We validate our approach by means of estimation and goodness-of-fit testing over simulated data, showing an accurate performance.
基金supported by National Natural Science Foundation of China (Grant Nos. 50805065, 51075179)
文摘Highly versatile machines, such as wheel loaders, forklifts, and mining haulers, are subject to many kinds of working conditions, as well as indefinite factors that lead to the complexity of the load. The load probability distribution function （PDF） of transmission gears has many distributions centers; thus, its PDF cannot be well represented by just a single-peak function. For the purpose of representing the distribution characteristics of the complicated phenomenon accurately, this paper proposes a novel method to establish a mixture model. Based on linear regression models and correlation coefficients, the proposed method can be used to automatically select the best-fitting function in the mixture model. Coefficient of determination, the mean square error, and the maximum deviation are chosen and then used as judging criteria to describe the fitting precision between the theoretical distribution and the corresponding histogram of the available load data. The applicability of this modeling method is illustrated by the field testing data of a wheel loader. Meanwhile, the load spectra based on the mixture model are compiled. The comparison results show that the mixture model is more suitable for the description of the load-distribution characteristics. The proposed research improves the flexibility and intelligence of modeling, reduces the statistical error and enhances the fitting accuracy, and the load spectra complied by this method can better reflect the actual load characteristic of the gear component.
文摘In this paper, we introduce a modification of the Quasi Lindley distribution which has various advantageous properties for the lifetime data. Several fundamental structural properties of the distribution are explored. Its density function can be left-skewed, symmetrical, and right-skewed shapes with various rages of tail-weights and dispersions. The failure rate function of the new dist</span><span style="font-family:Verdana;">ribution has the flexibility to be increasing, decreasing, constant, an</span><span style="font-family:Verdana;">d bathtub shapes. A simulation study is done to examine the performance of maximum likelihood and moment estimation methods in its unknown parameter estimations based on the asymptotic theory. The potentiality of the new distribution is illustrated by means of applications to the simulated and three real-world data sets.
文摘The fluctuation characteristics is the inherent property of wind power.Through analysis of a large number of wind t＇anns based on measured data,we find it describes the best probability distribution of wind power fluctuation for the mixed Gauss distribution of two components,and try to carry out the physical interpretation of two components.Further discussion is between the probability distribution of fluctuating wind power time difference and whole relationship.It is found that the two have basic similarity.Through comparing the different time level data quantified losses the information of wind power fluctuation,quantitative determination of the degree of impact prediction.We can summarize and understand of wind power fluctuation,constructing instance from the wind farm construction and monitoring prediction two aspect recommendations to overcome the adverse effects of wind power fluctuations on the power grid operation.
基金Supported by the National Natural Science Foundation of China (No. 61071091)Jiangsu Province Graduate Innovative Research Plan (CX07B_107Z)
文摘In Wyner-Ziv (WZ) Distributed Video Coding (DVC), correlation noise model is often used to describe the error distribution between WZ frame and the side information. The accuracy of the model can influence the performance of the video coder directly. A mixture correlation noise model in Discrete Cosine Transform (DCT) domain for WZ video coding is established in this paper. Different correlation noise estimation method is used for direct current and alternating current coefficients. Parameter estimation method based on expectation maximization algorithm is used to estimate the Laplace distribution center of direct current frequency band and Mixture Laplace-Uniform Distribution Model (MLUDM) is established for alternating current coefficients. Experimental results suggest that the proposed mixture correlation noise model can describe the heavy tail and sudden change of the noise accurately at high rate and make significant improvement on the coding efficiency compared with the noise model presented by DIStributed COding for Video sERvices (DISCOVER).
基金Funded by the National Natural Science Foundation of China(No.51578290)
文摘In order to analyze the microstructure of salt anti-freezing asphalt concrete, i e, MFL（Mafilon） modified asphalt concrete, MIP（mercury intrusion porosity） method was used to obtain the data including porosity and pore size distribution in micro scale. Results show that the porosity grows up with the increase of immersion duration and the salt content. During the immersion, the amount of large pores（60-200 μm） grow up gradually and porosity also grows up correspondingly. Even with different immersion duration, most pores＇ size distribute is beyond 7000 nm.
基金Supported by the National Natural Science Foundation of China(Grant No.61304155)Beijing Municipal Party Committee Organization Department Talents Project(Grant No.2012D005003000005)Graduate Department of BTBU Comprehensive Reform Project to Promote Talent Cultivation(Grant No.19005428069)
文摘A new family of univariate exponential slash distribution is introduced, which is based on elliptical distributions and defined by means of a stochastic representation as the scale mixture of an elliptically distributed random variable with respect to the power of an exponential random variable. The same idea is extended to the multivariate case. General properties of the resulting families, including their moments and kurtosis coefficient, are stud- ied. And inferences based on methods of moment and maximum likelihood are discussed. A real data is presented to show this family is flexible and fits much better than other related families.
文摘The classical risk process that is perturbed by diffusion is studied. The explicit expressions for the ruin probability and the surplus distribution of the risk process at the time of ruin are obtained when the claim amount distribution is a finite mixture of exponential distributions or a Gamma (2, α) distribution.
文摘This paper compares the stock return distribution models of mixture normal distribution, mixed diffusion-jump and GARCH models based on the data of Chinese stock market. The Schwarz criterion is also used. We find all these models can capture the features of stock returns partly. EGARCH model is the best fitting to daily return and stable during different period. When the weekly and monthly returns are tested, the differences of the models＇ fitness become unobvious and unstable.
基金the National Natural Science Foundation of China(11861041,11261025).
文摘Mixture of Experts(MoE)regression models are widely studied in statistics and machine learning for modeling heterogeneity in data for regression,clustering and classification.Laplace distribution is one of the most important statistical tools to analyze thick and tail data.Laplace Mixture of Linear Experts(LMoLE)regression models are based on the Laplace distribution which is more robust.Similar to modelling variance parameter in a homogeneous population,we propose and study a new novel class of models:heteroscedastic Laplace mixture of experts regression models to analyze the heteroscedastic data coming from a heterogeneous population in this paper.The issues of maximum likelihood estimation are addressed.In particular,Minorization-Maximization(MM)algorithm for estimating the regression parameters is developed.Properties of the estimators of the regression coefficients are evaluated through Monte Carlo simulations.Results from the analysis of two real data sets are presented.
文摘Performance of Turbo-Codes in communication channels with impulsive noise is analyzed. First, mathematical model of impulsive noise is presented because it has non-Gaussian nature and is found in many wireless channels due to impulsive phenomena of radio-frequency interference. Then, with linear Log-MAP decoding algorithm for its low complexity, Turbo-Codes are adopted and analyzed in such communication channels. To confirm the performance of the proposed method, simulations on both static and fully interleaved flat Rayleigh fading channels with impulsive noise have been carried out. It is shown that Turbo-Codes have a better performance than the conventional methods （e.g. convolutionally coded system）.
文摘The major male sterile genes in a new photo/thermo-sensitive genie male sterile (PTGMS) line B06S of rice were analyzed by the manipulation of mixture distribution theory. The results indicated that a pair of major male sterile nuclear genes with large effects were responsible for controlling the male sterility of B06S.
文摘The paper presents an algorithm of automatic target detection in Synthetic Aperture Radar(SAR) images based on Maximum A Posteriori(MAP). The algorithm is divided into three steps. First, it employs Gaussian mixture distribution to approximate and estimate multi-modal histogram of SAR image. Then, based on the principle of MAP, when a priori probability is both unknown and learned respectively, the sample pixels are classified into different classes c = {target,shadow, background}. Last, it compares the results of two different target detections. Simulation results preferably indicate that the presented algorithm is fast and robust, with the learned a priori probability, an approach to target detection is reliable and promising.
文摘Generalized method of moments based on probability generating function is considered. Estimation and model testing are unified using this approach which also leads to distribution free chi-square tests. The estimation methods developed are also related to estimation methods based on generalized estimating equations but with the advantage of having statistics for model testing. The methods proposed overcome numerical problems often encountered when the probability mass functions have no closed forms which prevent the use of maximum likelihood (ML) procedures and in general, ML procedures do not lead to distribution free model testing statistics.
文摘Hybrid clustering combines partitional and hierarchical clustering for computational effectiveness and versatility in cluster shape. In such clustering, a dissimilarity measure plays a crucial role in the hierarchical merging. The dissimilarity measure has great impact on the final clustering, and data-independent properties are needed to choose the right dissimilarity measure for the problem at hand. Properties for distance-based dissimilarity measures have been studied for decades, but properties for density-based dissimilarity measures have so far received little attention. Here, we propose six data-independent properties to evaluate density-based dissimilarity measures associated with hybrid clustering, regarding equality, orthogonality, symmetry, outlier and noise observations, and light-tailed models for heavy-tailed clusters. The significance of the properties is investigated, and we study some well-known dissimilarity measures based on Shannon entropy, misclassification rate, Bhattacharyya distance and Kullback-Leibler divergence with respect to the proposed properties. As none of them satisfy all the proposed properties, we introduce a new dissimilarity measure based on the Kullback-Leibler information and show that it satisfies all proposed properties. The effect of the proposed properties is also illustrated on several real and simulated data sets.
