With the increasing availability of precipitation radar data from space,enhancement of the resolution of spaceborne precipitation observations is important,particularly for hazard prediction and climate modeling at lo...With the increasing availability of precipitation radar data from space,enhancement of the resolution of spaceborne precipitation observations is important,particularly for hazard prediction and climate modeling at local scales relevant to extreme precipitation intensities and gradients.In this paper,the statistical characteristics of radar precipitation reflectivity data are studied and modeled using a hidden Markov tree(HMT)in the wavelet domain.Then,a high-resolution interpolation algorithm is proposed for spaceborne radar reflectivity using the HMT model as prior information.Owing to the small and transient storm elements embedded in the larger and slowly varying elements,the radar precipitation data exhibit distinct multiscale statistical properties,including a non-Gaussian structure and scale-to-scale dependency.An HMT model can capture well the statistical properties of radar precipitation,where the wavelet coefficients in each sub-band are characterized as a Gaussian mixture model(GMM),and the wavelet coefficients from the coarse scale to fine scale are described using a multiscale Markov process.The state probabilities of the GMM are determined using the expectation maximization method,and other parameters,for instance,the variance decay parameters in the HMT model are learned and estimated from high-resolution ground radar reflectivity images.Using the prior model,the wavelet coefficients at finer scales are estimated using local Wiener filtering.The interpolation algorithm is validated using data from the precipitation radar onboard the Tropical Rainfall Measurement Mission satellite,and the reconstructed results are found to be able to enhance the spatial resolution while optimally reproducing the local extremes and gradients.展开更多
We present a factorial representation of Gaussian mixture models for observation densities in Hidden Markov Models(HMMs), which uses the factorial learning in the HMM framework. We derive the reestimation formulas for...We present a factorial representation of Gaussian mixture models for observation densities in Hidden Markov Models(HMMs), which uses the factorial learning in the HMM framework. We derive the reestimation formulas for estimating the factorized parameters by the Expectation Maximization (EM) algorithm. We conduct several experiments to compare the performance of this model structure with Factorial Hidden Markov Models(FHMMs) and HMMs, some conclusions and promising empirical results are presented.展开更多
Purpose:This study aims to provide the best estimate of a stock’s next day’s closing price for a given day with the help of the hidden Markov model:Gaussian mixture model(HMM-GMM).The results were compared with Hass...Purpose:This study aims to provide the best estimate of a stock’s next day’s closing price for a given day with the help of the hidden Markov model:Gaussian mixture model(HMM-GMM).The results were compared with Hassan and Nath’s(2005)study using HMM and artificial neural network(ANN).Design/methodology/approach:The study adopted an initialization approach wherein the hidden states of the HMM are modelled as GMM using two different approaches.Training of the HMM-GMM model is carried out using two methods.The prediction was performed by taking the closest closing price(having a loglikelihood within the tolerance range)to that of the present one as the closing price for the next day.Mean absolute percentage error(MAPE)has been used to compare the proposed GMM-HMM model against the models of the research study(Hassan and Nath,2005).Findings:Comparing this study with Hassan and Nath(2005)reveals that the proposed model outperformed in 66 out of the 72 different test cases.The results affirm that the model can be used for more accurate time series prediction.Further,compared with the results of the ANN model from Hassan’s study,the proposed HMM model outperformed 24 of the 36 test cases.Originality/value:The study introduced a novel initialization and two training/prediction approaches for the HMM-GMM model.It is to be noted that the study has introduced a GMM-HMM-based closing price estimator for stock price prediction.The proposed method of forecasting the stock prices using GMM-HMM is explainable and has a solid statistical foundation.展开更多
In this paper,we present a comparison of Khasi speech representations with four different spectral features and novel extension towards the development of Khasi speech corpora.These four features include linear predic...In this paper,we present a comparison of Khasi speech representations with four different spectral features and novel extension towards the development of Khasi speech corpora.These four features include linear predictive coding(LPC),linear prediction cepstrum coefficient(LPCC),perceptual linear prediction(PLP),and Mel frequency cepstral coefficient(MFCC).The 10-hour speech data were used for training and 3-hour data for testing.For each spectral feature,different hidden Markov model(HMM)based recognizers with variations in HMM states and different Gaussian mixture models(GMMs)were built.The performance was evaluated by using the word error rate(WER).The experimental results show that MFCC provides a better representation for Khasi speech compared with the other three spectral features.展开更多
The authors propose a two-stage method for recognizing driving situations on the basis of driving signals for application to a safe human interface of an in-vehicle information system. In first stage, an unknown drivi...The authors propose a two-stage method for recognizing driving situations on the basis of driving signals for application to a safe human interface of an in-vehicle information system. In first stage, an unknown driving situation is determined as stopping behavior or non-stopping behavior. In second stage, a Hidden Markov Model （HMM）-based pattern recognition method is used to model and recognize six non-stopping driving situations. The authors attempt to find the optimal HMM configuration to improve the performance of driving situation recognition. Center for Integrated Acoustic Information Research （CLAIR） in-vehicle corpus is used to evaluate the HMM-based recognition method. Driving situation categories are recognized using five driving signals. The proposed method achieves a relative error reduction rate of 30.9% compared to a conventional one-stage based HMMs.展开更多
随着风电大量并入电力系统,能源储运协调规划越来越重要。针对目前面向需求响应的能源规划方法不能准确描述风电场出力时间序列且难以达到最优解的问题,引入了一种将高斯混合模型和隐马尔可夫模型的统计模型。与传统隐马尔可夫模型相比...随着风电大量并入电力系统,能源储运协调规划越来越重要。针对目前面向需求响应的能源规划方法不能准确描述风电场出力时间序列且难以达到最优解的问题,引入了一种将高斯混合模型和隐马尔可夫模型的统计模型。与传统隐马尔可夫模型相比,该模型不仅能捕捉数据的时序特性,还能通过高斯混合模型更灵活地描述数据的概率分布。基于该模型设计了一种风电场功率输出数据的生成方法,并采用乔莱斯基分解算法将序列转化为源荷不确定集合,设计了面向需求响应的能源储运协调规划方法,将风电出力时间序列生成方法与其他方法进行对比分析,并将规划方法以改进的IEEE Reliability Test System 1979系统为例进行验证。结果表明,该方法的最高均方误差和平均绝对误差分别为0.089和0.221,均低于对比方法;研究提出的方法具有有效性并能较好地实现最优规划。研究结果可为能源储运协调规划相关研究提供借鉴。展开更多
为满足智能车辆的个性化需求,提高智能车辆人-机交互协同的满意度和接受度,构筑双层驾驶人跟驰模型框架,提出自适应驾驶人期望跟车间距和行为习惯的个性化驾驶人跟驰模型。首先,提取个体驾驶人跟驰均衡状态的数据,采用高斯混合和概率密...为满足智能车辆的个性化需求,提高智能车辆人-机交互协同的满意度和接受度,构筑双层驾驶人跟驰模型框架,提出自适应驾驶人期望跟车间距和行为习惯的个性化驾驶人跟驰模型。首先,提取个体驾驶人跟驰均衡状态的数据,采用高斯混合和概率密度函数(Gaussian Mixture Model and Probability Density Function,GMM-PDF)建立第1层模型,即驾驶人期望跟车距离模型。然后,将期望跟车距离参数引入模型,基于高斯混合-隐马尔可夫方法(Gaussian Mixture Model and Hidden Markov Model,GMM-HMM)学习驾驶习性,建立第2层模型预测加速度,即个性化驾驶人跟驰模型。其次,研究不同高斯分量个数对模型效果的影响,对比双层模型与Gipps模型、最优间距模型(Optimal Distance Model,ODM)、单层模型及通用模型的性能。最后,8位被试驾驶人的自然驾驶行为数据验证结果表明:高斯分量数量与模型性能存在一定的正相关性;在最优高斯分量数量下,8位被试驾驶人在训练集上预测误差均值为0.101 m·s^(-2),在测试集上为0.123 m·s^(-2);随机选取其中1位驾驶人的2个跟车片段数据进行模型计算,结果显示,加速度的平均误差绝对值分别为0.087 m·s^(-2)和0.096 m·s^(-2),预测效果优于Gipps模型、ODM模型、单层模型及通用模型30%以上,与驾驶人实际跟驰行为的吻合度更高。展开更多
基金This study was funded by the National Natural Science Foundation of China(Grant No.41975027)the Natural Science Foundation of Jiangsu Province(Grant No.BK20171457)the National Key R&D Program on Monitoring,Early Warning and Prevention of Major Natural Disasters(Grant No.2017YFC1501401).
文摘With the increasing availability of precipitation radar data from space,enhancement of the resolution of spaceborne precipitation observations is important,particularly for hazard prediction and climate modeling at local scales relevant to extreme precipitation intensities and gradients.In this paper,the statistical characteristics of radar precipitation reflectivity data are studied and modeled using a hidden Markov tree(HMT)in the wavelet domain.Then,a high-resolution interpolation algorithm is proposed for spaceborne radar reflectivity using the HMT model as prior information.Owing to the small and transient storm elements embedded in the larger and slowly varying elements,the radar precipitation data exhibit distinct multiscale statistical properties,including a non-Gaussian structure and scale-to-scale dependency.An HMT model can capture well the statistical properties of radar precipitation,where the wavelet coefficients in each sub-band are characterized as a Gaussian mixture model(GMM),and the wavelet coefficients from the coarse scale to fine scale are described using a multiscale Markov process.The state probabilities of the GMM are determined using the expectation maximization method,and other parameters,for instance,the variance decay parameters in the HMT model are learned and estimated from high-resolution ground radar reflectivity images.Using the prior model,the wavelet coefficients at finer scales are estimated using local Wiener filtering.The interpolation algorithm is validated using data from the precipitation radar onboard the Tropical Rainfall Measurement Mission satellite,and the reconstructed results are found to be able to enhance the spatial resolution while optimally reproducing the local extremes and gradients.
文摘We present a factorial representation of Gaussian mixture models for observation densities in Hidden Markov Models(HMMs), which uses the factorial learning in the HMM framework. We derive the reestimation formulas for estimating the factorized parameters by the Expectation Maximization (EM) algorithm. We conduct several experiments to compare the performance of this model structure with Factorial Hidden Markov Models(FHMMs) and HMMs, some conclusions and promising empirical results are presented.
文摘Purpose:This study aims to provide the best estimate of a stock’s next day’s closing price for a given day with the help of the hidden Markov model:Gaussian mixture model(HMM-GMM).The results were compared with Hassan and Nath’s(2005)study using HMM and artificial neural network(ANN).Design/methodology/approach:The study adopted an initialization approach wherein the hidden states of the HMM are modelled as GMM using two different approaches.Training of the HMM-GMM model is carried out using two methods.The prediction was performed by taking the closest closing price(having a loglikelihood within the tolerance range)to that of the present one as the closing price for the next day.Mean absolute percentage error(MAPE)has been used to compare the proposed GMM-HMM model against the models of the research study(Hassan and Nath,2005).Findings:Comparing this study with Hassan and Nath(2005)reveals that the proposed model outperformed in 66 out of the 72 different test cases.The results affirm that the model can be used for more accurate time series prediction.Further,compared with the results of the ANN model from Hassan’s study,the proposed HMM model outperformed 24 of the 36 test cases.Originality/value:The study introduced a novel initialization and two training/prediction approaches for the HMM-GMM model.It is to be noted that the study has introduced a GMM-HMM-based closing price estimator for stock price prediction.The proposed method of forecasting the stock prices using GMM-HMM is explainable and has a solid statistical foundation.
基金supported by the Visvesvaraya Ph.D.Scheme for Electronics and IT students launched by the Ministry of Electronics and Information Technology(MeiTY),Government of India under Grant No.PhD-MLA/4(95)/2015-2016.
文摘In this paper,we present a comparison of Khasi speech representations with four different spectral features and novel extension towards the development of Khasi speech corpora.These four features include linear predictive coding(LPC),linear prediction cepstrum coefficient(LPCC),perceptual linear prediction(PLP),and Mel frequency cepstral coefficient(MFCC).The 10-hour speech data were used for training and 3-hour data for testing.For each spectral feature,different hidden Markov model(HMM)based recognizers with variations in HMM states and different Gaussian mixture models(GMMs)were built.The performance was evaluated by using the word error rate(WER).The experimental results show that MFCC provides a better representation for Khasi speech compared with the other three spectral features.
文摘The authors propose a two-stage method for recognizing driving situations on the basis of driving signals for application to a safe human interface of an in-vehicle information system. In first stage, an unknown driving situation is determined as stopping behavior or non-stopping behavior. In second stage, a Hidden Markov Model （HMM）-based pattern recognition method is used to model and recognize six non-stopping driving situations. The authors attempt to find the optimal HMM configuration to improve the performance of driving situation recognition. Center for Integrated Acoustic Information Research （CLAIR） in-vehicle corpus is used to evaluate the HMM-based recognition method. Driving situation categories are recognized using five driving signals. The proposed method achieves a relative error reduction rate of 30.9% compared to a conventional one-stage based HMMs.
文摘随着风电大量并入电力系统,能源储运协调规划越来越重要。针对目前面向需求响应的能源规划方法不能准确描述风电场出力时间序列且难以达到最优解的问题,引入了一种将高斯混合模型和隐马尔可夫模型的统计模型。与传统隐马尔可夫模型相比,该模型不仅能捕捉数据的时序特性,还能通过高斯混合模型更灵活地描述数据的概率分布。基于该模型设计了一种风电场功率输出数据的生成方法,并采用乔莱斯基分解算法将序列转化为源荷不确定集合,设计了面向需求响应的能源储运协调规划方法,将风电出力时间序列生成方法与其他方法进行对比分析,并将规划方法以改进的IEEE Reliability Test System 1979系统为例进行验证。结果表明,该方法的最高均方误差和平均绝对误差分别为0.089和0.221,均低于对比方法;研究提出的方法具有有效性并能较好地实现最优规划。研究结果可为能源储运协调规划相关研究提供借鉴。
文摘为满足智能车辆的个性化需求,提高智能车辆人-机交互协同的满意度和接受度,构筑双层驾驶人跟驰模型框架,提出自适应驾驶人期望跟车间距和行为习惯的个性化驾驶人跟驰模型。首先,提取个体驾驶人跟驰均衡状态的数据,采用高斯混合和概率密度函数(Gaussian Mixture Model and Probability Density Function,GMM-PDF)建立第1层模型,即驾驶人期望跟车距离模型。然后,将期望跟车距离参数引入模型,基于高斯混合-隐马尔可夫方法(Gaussian Mixture Model and Hidden Markov Model,GMM-HMM)学习驾驶习性,建立第2层模型预测加速度,即个性化驾驶人跟驰模型。其次,研究不同高斯分量个数对模型效果的影响,对比双层模型与Gipps模型、最优间距模型(Optimal Distance Model,ODM)、单层模型及通用模型的性能。最后,8位被试驾驶人的自然驾驶行为数据验证结果表明:高斯分量数量与模型性能存在一定的正相关性;在最优高斯分量数量下,8位被试驾驶人在训练集上预测误差均值为0.101 m·s^(-2),在测试集上为0.123 m·s^(-2);随机选取其中1位驾驶人的2个跟车片段数据进行模型计算,结果显示,加速度的平均误差绝对值分别为0.087 m·s^(-2)和0.096 m·s^(-2),预测效果优于Gipps模型、ODM模型、单层模型及通用模型30%以上,与驾驶人实际跟驰行为的吻合度更高。
