High pressure die casting (HPDC) is a versatile material processing method for mass-production of metal parts with complex geometries,and this method has been widely used in manufacturing various products of excellent...High pressure die casting (HPDC) is a versatile material processing method for mass-production of metal parts with complex geometries,and this method has been widely used in manufacturing various products of excellent dimensional accuracy and productivity. In order to ensure the quality of the components,a number of variables need to be properly set. A novel methodology for high pressure die casting process optimization was developed,validated and applied to selection of optimal parameters,which incorporate design of experiment (DOE),Gaussian process (GP) regression technique and genetic algorithms (GA). This new approach was applied to process optimization for cast magnesium alloy notebook shell. After being trained,using data generated by PROCAST (FEM-based simulation software),the GP model approximated well with the simulation by extracting useful information from the simulation results. With the help of MATLAB,the GP/GA based approach has achieved the optimum solution of die casting process condition settings.展开更多
The EM algorithm is a very popular maximum likelihood estimation method, the iterative algorithm for solving the maximum likelihood estimator when the observation data is the incomplete data, but also is very effectiv...The EM algorithm is a very popular maximum likelihood estimation method, the iterative algorithm for solving the maximum likelihood estimator when the observation data is the incomplete data, but also is very effective algorithm to estimate the finite mixture model parameters. However, EM algorithm can not guarantee to find the global optimal solution, and often easy to fall into local optimal solution, so it is sensitive to the determination of initial value to iteration. Traditional EM algorithm select the initial value at random, we propose an improved method of selection of initial value. First, we use the k-nearest-neighbor method to delete outliers. Second, use the k-means to initialize the EM algorithm. Compare this method with the original random initial value method, numerical experiments show that the parameter estimation effect of the initialization of the EM algorithm is significantly better than the effect of the original EM algorithm.展开更多
The key problem of the adaptive mixture background model is that the parameters can adaptively change according to the input data. To address the problem, a new method is proposed. Firstly, the recursive equations are...The key problem of the adaptive mixture background model is that the parameters can adaptively change according to the input data. To address the problem, a new method is proposed. Firstly, the recursive equations are inferred based on the maximum likelihood rule. Secondly, the forgetting factor and learning rate factor are redefined, and their still more general formulations are obtained by analyzing their practical functions. Lastly, the convergence of the proposed algorithm is proved to enable the estimation converge to a local maximum of the data likelihood function according to the stochastic approximation theory. The experiments show that the proposed learning algorithm excels the formers both in converging rate and accuracy.展开更多
With the increment of the number of Gaussian components, the computation cost increases in the Gaussian mixture probability hypothesis density(GM-PHD) filter. Based on the theory of Chen et al, we propose an improved ...With the increment of the number of Gaussian components, the computation cost increases in the Gaussian mixture probability hypothesis density(GM-PHD) filter. Based on the theory of Chen et al, we propose an improved pruning algorithm for the GM-PHD filter, which utilizes not only the Gaussian components’ means and covariance, but their weights as a new criterion to improve the estimate accuracy of the conventional pruning algorithm for tracking very closely proximity targets. Moreover, it solves the end-less while-loop problem without the need of a second merging step. Simulation results show that this improved algorithm is easier to implement and more robust than the formal ones.展开更多
An efficient approach was proposed for discriminating shadows from moving objects. In the background subtraction stage, moving objects were extracted. Then, the initial classification for moving shadow pixels and fore...An efficient approach was proposed for discriminating shadows from moving objects. In the background subtraction stage, moving objects were extracted. Then, the initial classification for moving shadow pixels and foreground object pixels was performed by using color invariant features. In the shadow model learning stage, instead of a single Gaussian distribution, it was assumed that the density function computed on the values of chromaticity difference or bright difference, can be modeled as a mixture of Gaussian consisting of two density functions. Meanwhile, the Gaussian parameter estimation was performed by using EM algorithm. The estimates were used to obtain shadow mask according to two constraints. Finally, experiments were carried out. The visual experiment results confirm the effectiveness of proposed method. Quantitative results in terms of the shadow detection rate and the shadow discrimination rate(the maximum values are 85.79% and 97.56%, respectively) show that the proposed approach achieves a satisfying result with post-processing step.展开更多
To identify systems with non-uniformly sampled input data, a recursive Bayesian identification algorithm with covariance resetting is proposed. Using estimated noise transfer function as a dynamic filter, the system w...To identify systems with non-uniformly sampled input data, a recursive Bayesian identification algorithm with covariance resetting is proposed. Using estimated noise transfer function as a dynamic filter, the system with colored noise is transformed into the system with white noise. In order to improve estimates, the estimated noise variance is employed as a weighting factor in the algorithm. Meanwhile, a modified covariance resetting method is also integrated in the proposed algorithm to increase the convergence rate. A numerical example and an industrial example validate the proposed algorithm.展开更多
In this paper, we study optimal recovery （reconstruction） of functions on the sphere in the average case setting. We obtain the asymptotic orders of average sampling numbers of a Sobolev space on the sphere with a G...In this paper, we study optimal recovery （reconstruction） of functions on the sphere in the average case setting. We obtain the asymptotic orders of average sampling numbers of a Sobolev space on the sphere with a Gaussian measure in the Lq （S^d-1） metric for 1 ≤ q ≤ ∞, and show that some worst-case asymptotically optimal algorithms are also asymptotically optimal in the average case setting in the Lq （S^d-1） metric for 1 ≤ q ≤ ∞.展开更多
Detailed mathematical deduction is presented to improve the graphics quality of aircraft dashboards. Generation of rhumb lines with such kind of algorithm shows that the new approach is effective.
The principle of digital watermark is the method of adding digital watermark in the frequency domain. The digital watermark hides the watermark in digital media, such as image, voice, video, etc., so as to realize the...The principle of digital watermark is the method of adding digital watermark in the frequency domain. The digital watermark hides the watermark in digital media, such as image, voice, video, etc., so as to realize the functions of copyright protection, and identity recognition. DCT for Discrete Cosine Transform is used to transform the image pixel value and the frequency domain coefficient matrix to realize the embedding and extracting of the blind watermark in the paper. After success, the image is attacked by white noise and Gaussian low-pass filtering. The result shows that the watermark signal embedded based on the DCT algorithm is relatively robust, and can effectively resist some attack methods that use signal distortion to destroy the watermark, and has good robustness and imperceptibility.展开更多
A genetic algorithm(GA)-based new method is designed to evaluate thecircularity error of mechanical parts. The method uses the capability of nonlinear optimization ofGA to search for the optimal solution of circularit...A genetic algorithm(GA)-based new method is designed to evaluate thecircularity error of mechanical parts. The method uses the capability of nonlinear optimization ofGA to search for the optimal solution of circularity error. The finely-designed GA (FDGA)characterized dynamical bisexual recombination and Gaussian mutation. The mathematical model of thenonlinear problem is given. The implementation details in FDGA are described such as the crossoveror recombination mechanism which utilized a bisexual reproduction scheme and the elitist reservationmethod; and the adaptive mutation which used the Gaussian probability distribution to determine thevalues of the offspring produced by mutation mechanism. The examples are provided to verify thedesigned FDGA. The computation results indicate that the FDGA works very well in the field of formerror evaluation such as circularity evaluation.展开更多
In this paper, a Gaussian mixture model (GMM) based classifier is described to tell whether precipitation events will happen on a certain day at a certain time from historical meteorological data. The classifier deals...In this paper, a Gaussian mixture model (GMM) based classifier is described to tell whether precipitation events will happen on a certain day at a certain time from historical meteorological data. The classifier deals with a two-class classification problem where one class represents precipitation events and the other represents non-precipitation events. The concept of ambiguity is introduced to represent cases where weather conditions between the two classes like drizzles, intermittent or overcast are more likely to happen. Six groups of experiments are carried out to evaluate the performance of the classifier using different configurations based on the observation data released by Shanghai Baoshan weather station. Specifically, a typical classification performance of about 75% accuracy, 30% precision and 80% recall is achieved for prediction tasks with a time span of 12 hours.展开更多
In this paper,we make use of the boosting method to introduce a new learning algorithm for Gaussian Mixture Models (GMMs) called adapted Boosted Mixture Learning (BML). The method possesses the ability to rectify the ...In this paper,we make use of the boosting method to introduce a new learning algorithm for Gaussian Mixture Models (GMMs) called adapted Boosted Mixture Learning (BML). The method possesses the ability to rectify the existing problems in other conventional techniques for estimating the GMM parameters, due in part to a new mixing-up strategy to increase the number of Gaussian components. The discriminative splitting idea is employed for Gaussian mixture densities followed by learning via the introduced method. Then, the GMM classifier was applied to distinguish between healthy infants and those that present a selected set of medical conditions. Each group includes both full-term and premature infants. Cry-pattern for each pathological condition is created by using the adapted BML method and 13-dimensional Mel-Frequency Cepstral Coefficients (MFCCs) feature vector. The test results demonstrate that the introduced method for training GMMs has a better performance than the traditional method based upon random splitting and EM-based re-estimation as a reference system in multi-pathological classification task.展开更多
文摘High pressure die casting (HPDC) is a versatile material processing method for mass-production of metal parts with complex geometries,and this method has been widely used in manufacturing various products of excellent dimensional accuracy and productivity. In order to ensure the quality of the components,a number of variables need to be properly set. A novel methodology for high pressure die casting process optimization was developed,validated and applied to selection of optimal parameters,which incorporate design of experiment (DOE),Gaussian process (GP) regression technique and genetic algorithms (GA). This new approach was applied to process optimization for cast magnesium alloy notebook shell. After being trained,using data generated by PROCAST (FEM-based simulation software),the GP model approximated well with the simulation by extracting useful information from the simulation results. With the help of MATLAB,the GP/GA based approach has achieved the optimum solution of die casting process condition settings.
文摘The EM algorithm is a very popular maximum likelihood estimation method, the iterative algorithm for solving the maximum likelihood estimator when the observation data is the incomplete data, but also is very effective algorithm to estimate the finite mixture model parameters. However, EM algorithm can not guarantee to find the global optimal solution, and often easy to fall into local optimal solution, so it is sensitive to the determination of initial value to iteration. Traditional EM algorithm select the initial value at random, we propose an improved method of selection of initial value. First, we use the k-nearest-neighbor method to delete outliers. Second, use the k-means to initialize the EM algorithm. Compare this method with the original random initial value method, numerical experiments show that the parameter estimation effect of the initialization of the EM algorithm is significantly better than the effect of the original EM algorithm.
基金the Doctorate Foundation of the Engineering College, Air Force Engineering University.
文摘The key problem of the adaptive mixture background model is that the parameters can adaptively change according to the input data. To address the problem, a new method is proposed. Firstly, the recursive equations are inferred based on the maximum likelihood rule. Secondly, the forgetting factor and learning rate factor are redefined, and their still more general formulations are obtained by analyzing their practical functions. Lastly, the convergence of the proposed algorithm is proved to enable the estimation converge to a local maximum of the data likelihood function according to the stochastic approximation theory. The experiments show that the proposed learning algorithm excels the formers both in converging rate and accuracy.
基金supported by the National Natural Science Foundation of China(61703228)
文摘With the increment of the number of Gaussian components, the computation cost increases in the Gaussian mixture probability hypothesis density(GM-PHD) filter. Based on the theory of Chen et al, we propose an improved pruning algorithm for the GM-PHD filter, which utilizes not only the Gaussian components’ means and covariance, but their weights as a new criterion to improve the estimate accuracy of the conventional pruning algorithm for tracking very closely proximity targets. Moreover, it solves the end-less while-loop problem without the need of a second merging step. Simulation results show that this improved algorithm is easier to implement and more robust than the formal ones.
基金Project(50805023)supported by the National Natural Science Foundation of ChinaProject(BA2010093)supported by the Special Fund of Jiangsu Province for the Transformation of Scientific and Technological Achievements,ChinaProject(2008144)supported by the Hexa-type Elites Peak Program of Jiangsu Province,China
文摘An efficient approach was proposed for discriminating shadows from moving objects. In the background subtraction stage, moving objects were extracted. Then, the initial classification for moving shadow pixels and foreground object pixels was performed by using color invariant features. In the shadow model learning stage, instead of a single Gaussian distribution, it was assumed that the density function computed on the values of chromaticity difference or bright difference, can be modeled as a mixture of Gaussian consisting of two density functions. Meanwhile, the Gaussian parameter estimation was performed by using EM algorithm. The estimates were used to obtain shadow mask according to two constraints. Finally, experiments were carried out. The visual experiment results confirm the effectiveness of proposed method. Quantitative results in terms of the shadow detection rate and the shadow discrimination rate(the maximum values are 85.79% and 97.56%, respectively) show that the proposed approach achieves a satisfying result with post-processing step.
基金supported by National Natural Science Foundation of China(Nos.61273142 and 51477070)the Priority Academic Program Development of Jiangsu Higher Education Institutions(PAPD)Foundation for Six Talents by Jiangsu Province and Graduate Scientific Innovation Projects of Jiangsu University(No.KYXX_0003)
文摘To identify systems with non-uniformly sampled input data, a recursive Bayesian identification algorithm with covariance resetting is proposed. Using estimated noise transfer function as a dynamic filter, the system with colored noise is transformed into the system with white noise. In order to improve estimates, the estimated noise variance is employed as a weighting factor in the algorithm. Meanwhile, a modified covariance resetting method is also integrated in the proposed algorithm to increase the convergence rate. A numerical example and an industrial example validate the proposed algorithm.
基金supported by the National Natural Science Foundation of China(No.11426179)the National Natural Science Foundation of China(Nos.10871132,11271263)+4 种基金the Key Scientific Research Fund of Xihua University(No.z1312624)the Foundation of Sichuan Educational Committee(No.14ZA0112)the Preeminent Youth Fund for School of Science in Xihua Universitythe Beijing Natural Science Foundation(No.1132001)BCMIIS
文摘In this paper, we study optimal recovery （reconstruction） of functions on the sphere in the average case setting. We obtain the asymptotic orders of average sampling numbers of a Sobolev space on the sphere with a Gaussian measure in the Lq （S^d-1） metric for 1 ≤ q ≤ ∞, and show that some worst-case asymptotically optimal algorithms are also asymptotically optimal in the average case setting in the Lq （S^d-1） metric for 1 ≤ q ≤ ∞.
基金Supported by the condition guarantee of 995 engineering (sx004)
文摘Detailed mathematical deduction is presented to improve the graphics quality of aircraft dashboards. Generation of rhumb lines with such kind of algorithm shows that the new approach is effective.
文摘The principle of digital watermark is the method of adding digital watermark in the frequency domain. The digital watermark hides the watermark in digital media, such as image, voice, video, etc., so as to realize the functions of copyright protection, and identity recognition. DCT for Discrete Cosine Transform is used to transform the image pixel value and the frequency domain coefficient matrix to realize the embedding and extracting of the blind watermark in the paper. After success, the image is attacked by white noise and Gaussian low-pass filtering. The result shows that the watermark signal embedded based on the DCT algorithm is relatively robust, and can effectively resist some attack methods that use signal distortion to destroy the watermark, and has good robustness and imperceptibility.
基金The project is supported by National Natural Science Foundation of China(No.59975025).
文摘A genetic algorithm(GA)-based new method is designed to evaluate thecircularity error of mechanical parts. The method uses the capability of nonlinear optimization ofGA to search for the optimal solution of circularity error. The finely-designed GA (FDGA)characterized dynamical bisexual recombination and Gaussian mutation. The mathematical model of thenonlinear problem is given. The implementation details in FDGA are described such as the crossoveror recombination mechanism which utilized a bisexual reproduction scheme and the elitist reservationmethod; and the adaptive mutation which used the Gaussian probability distribution to determine thevalues of the offspring produced by mutation mechanism. The examples are provided to verify thedesigned FDGA. The computation results indicate that the FDGA works very well in the field of formerror evaluation such as circularity evaluation.
文摘In this paper, a Gaussian mixture model (GMM) based classifier is described to tell whether precipitation events will happen on a certain day at a certain time from historical meteorological data. The classifier deals with a two-class classification problem where one class represents precipitation events and the other represents non-precipitation events. The concept of ambiguity is introduced to represent cases where weather conditions between the two classes like drizzles, intermittent or overcast are more likely to happen. Six groups of experiments are carried out to evaluate the performance of the classifier using different configurations based on the observation data released by Shanghai Baoshan weather station. Specifically, a typical classification performance of about 75% accuracy, 30% precision and 80% recall is achieved for prediction tasks with a time span of 12 hours.
文摘In this paper,we make use of the boosting method to introduce a new learning algorithm for Gaussian Mixture Models (GMMs) called adapted Boosted Mixture Learning (BML). The method possesses the ability to rectify the existing problems in other conventional techniques for estimating the GMM parameters, due in part to a new mixing-up strategy to increase the number of Gaussian components. The discriminative splitting idea is employed for Gaussian mixture densities followed by learning via the introduced method. Then, the GMM classifier was applied to distinguish between healthy infants and those that present a selected set of medical conditions. Each group includes both full-term and premature infants. Cry-pattern for each pathological condition is created by using the adapted BML method and 13-dimensional Mel-Frequency Cepstral Coefficients (MFCCs) feature vector. The test results demonstrate that the introduced method for training GMMs has a better performance than the traditional method based upon random splitting and EM-based re-estimation as a reference system in multi-pathological classification task.
