Presents the abstract L...-norm error estimate of nonconforming finite element method. Use of the Aubin Nitsche Lemma in estimating nonconforming finite element methods; Details on the equations.
In this paper, we propose a new approach to the problem of degree reduction of Bézier curves based on the given endpoint constraints. A differential term is added for the purpose of controlling the smoothness to ...In this paper, we propose a new approach to the problem of degree reduction of Bézier curves based on the given endpoint constraints. A differential term is added for the purpose of controlling the smoothness to a certain extent. Considering the adjustment of second derivative in curve design, a modified objective function including two parts is constructed here. One part is a kind of measure of the distance between original high order Bézier curve and degree-reduced curve. The other part represents the second derivative of degree-reduced curve. We tackle two kinds of conditions which are position vector constraint and tangent vector constraint respectively. The explicit representations of unknown points are presented. Some examples are illustrated to show the influence of the differential terms to approximation and smoothness effect.展开更多
Echo canceller generally needs a double-talk detector which is used to keep the adaptive filter from diverging in the appearance of near-end speech. In this paper we adopt a new double-talk detection algorithm based o...Echo canceller generally needs a double-talk detector which is used to keep the adaptive filter from diverging in the appearance of near-end speech. In this paper we adopt a new double-talk detection algorithm based onl 2 norm to detect the existence of near-end speech in an acoustic echo canceller. We analyze this algorithm from the point of view of functional analysis and point out that the proposed double-talk detection algorithm has the same performance as the classic one in a finite Banach space. The remarkable feature of this algorithm is its higher accuracy and better computation complexity. The fine properties of this algorithm are confirmed by computer simulation and the application in a multimedia communication system. Key words acoustic echo cancellation - double-talk, detection - l 2 norm - adaptive FIR CLC number TN 911 Foundation item: Supported by the the National High Technology Development of China (863-306-ZT05)Biography: Wang Shao-wei (1975-) male, Ph. D candidate, research direction: multimedia communication.展开更多
In this paper a mixed finite element-characteristic mixed finite element method is discussed to simulate an incompressible miscible Darcy-Forchheimer problem.The flow equation is solved by a mixed finite element and t...In this paper a mixed finite element-characteristic mixed finite element method is discussed to simulate an incompressible miscible Darcy-Forchheimer problem.The flow equation is solved by a mixed finite element and the approximation accuracy of Darch-Forchheimer velocity is improved one order.The concentration equation is solved by the method of mixed finite element,where the convection is discretized along the characteristic direction and the diffusion is discretized by the zero-order mixed finite element method.The characteristics can confirm strong stability at sharp fronts and avoids numerical dispersion and nonphysical oscillation.In actual computations the characteristics adopts a large time step without any loss of accuracy.The scalar unknowns and its adjoint vector function are obtained simultaneously and the law of mass conservation holds in every element by the zero-order mixed finite element discretization of diffusion flux.In order to derive the optimal 3/2-order error estimate in L^(2) norm,a post-processing technique is included in the approximation to the scalar unknowns.Numerical experiments are illustrated finally to validate theoretical analysis and efficiency.This method can be used to solve such an important problem.展开更多
面对网络图像的爆炸性增长,网络图像标注成为近年来一个热点研究内容,稀疏特征选择在提升网络图像标注效率和性能方面发挥着重要的作用.提出了一种增强稀疏性特征选择算法,即,基于l2,1/2矩阵范数和共享子空间的半监督稀疏特征选择算法(s...面对网络图像的爆炸性增长,网络图像标注成为近年来一个热点研究内容,稀疏特征选择在提升网络图像标注效率和性能方面发挥着重要的作用.提出了一种增强稀疏性特征选择算法,即,基于l2,1/2矩阵范数和共享子空间的半监督稀疏特征选择算法(semi-supervised sparse feature selection based on l2,1/2-matix norm with shared subspace learning,简称SFSLS)进行网络图像标注.在SFSLS算法中,应用l2,1/2矩阵范数来选取最稀疏和最具判别性的特征,通过共享子空间学习,考虑不同特征之间的关联信息.另外,基于图拉普拉斯的半监督学习,使SFSLS算法同时利用了有标签数据和无标签数据.设计了一种有效的迭代算法来最优化目标函数.SFSLS算法与其他稀疏特征选择算法在两个大规模网络图像数据库上进行了比较,结果表明,SFSLS算法更适合于大规模网络图像的标注.展开更多
文摘Presents the abstract L...-norm error estimate of nonconforming finite element method. Use of the Aubin Nitsche Lemma in estimating nonconforming finite element methods; Details on the equations.
文摘In this paper, we propose a new approach to the problem of degree reduction of Bézier curves based on the given endpoint constraints. A differential term is added for the purpose of controlling the smoothness to a certain extent. Considering the adjustment of second derivative in curve design, a modified objective function including two parts is constructed here. One part is a kind of measure of the distance between original high order Bézier curve and degree-reduced curve. The other part represents the second derivative of degree-reduced curve. We tackle two kinds of conditions which are position vector constraint and tangent vector constraint respectively. The explicit representations of unknown points are presented. Some examples are illustrated to show the influence of the differential terms to approximation and smoothness effect.
文摘Echo canceller generally needs a double-talk detector which is used to keep the adaptive filter from diverging in the appearance of near-end speech. In this paper we adopt a new double-talk detection algorithm based onl 2 norm to detect the existence of near-end speech in an acoustic echo canceller. We analyze this algorithm from the point of view of functional analysis and point out that the proposed double-talk detection algorithm has the same performance as the classic one in a finite Banach space. The remarkable feature of this algorithm is its higher accuracy and better computation complexity. The fine properties of this algorithm are confirmed by computer simulation and the application in a multimedia communication system. Key words acoustic echo cancellation - double-talk, detection - l 2 norm - adaptive FIR CLC number TN 911 Foundation item: Supported by the the National High Technology Development of China (863-306-ZT05)Biography: Wang Shao-wei (1975-) male, Ph. D candidate, research direction: multimedia communication.
基金supported by the Natural ScienceFoundation of Shandong Province(ZR2021MA019)。
文摘In this paper a mixed finite element-characteristic mixed finite element method is discussed to simulate an incompressible miscible Darcy-Forchheimer problem.The flow equation is solved by a mixed finite element and the approximation accuracy of Darch-Forchheimer velocity is improved one order.The concentration equation is solved by the method of mixed finite element,where the convection is discretized along the characteristic direction and the diffusion is discretized by the zero-order mixed finite element method.The characteristics can confirm strong stability at sharp fronts and avoids numerical dispersion and nonphysical oscillation.In actual computations the characteristics adopts a large time step without any loss of accuracy.The scalar unknowns and its adjoint vector function are obtained simultaneously and the law of mass conservation holds in every element by the zero-order mixed finite element discretization of diffusion flux.In order to derive the optimal 3/2-order error estimate in L^(2) norm,a post-processing technique is included in the approximation to the scalar unknowns.Numerical experiments are illustrated finally to validate theoretical analysis and efficiency.This method can be used to solve such an important problem.
文摘面对网络图像的爆炸性增长,网络图像标注成为近年来一个热点研究内容,稀疏特征选择在提升网络图像标注效率和性能方面发挥着重要的作用.提出了一种增强稀疏性特征选择算法,即,基于l2,1/2矩阵范数和共享子空间的半监督稀疏特征选择算法(semi-supervised sparse feature selection based on l2,1/2-matix norm with shared subspace learning,简称SFSLS)进行网络图像标注.在SFSLS算法中,应用l2,1/2矩阵范数来选取最稀疏和最具判别性的特征,通过共享子空间学习,考虑不同特征之间的关联信息.另外,基于图拉普拉斯的半监督学习,使SFSLS算法同时利用了有标签数据和无标签数据.设计了一种有效的迭代算法来最优化目标函数.SFSLS算法与其他稀疏特征选择算法在两个大规模网络图像数据库上进行了比较,结果表明,SFSLS算法更适合于大规模网络图像的标注.
