期刊文献+
共找到15篇文章
< 1 >
每页显示 20 50 100
基于核双子空间线性判别分析人脸识别方法
1
作者 葛熠 王亭亭 +1 位作者 韩月 李峰 《科技视界》 2012年第23期222-224,228,共4页
基于Fishers判别准则的LDA方法在人脸识别中处理高维数据时,通常会遇到欠采样的问题,且稳定性欠缺。本文提出的核双子线性判别分析方法(KDS-DA),通过对LDA、DS-LDA、KDS-DA的层层深入研究,在双子空间中,将核方法与线性判别分析(LDA)方... 基于Fishers判别准则的LDA方法在人脸识别中处理高维数据时,通常会遇到欠采样的问题,且稳定性欠缺。本文提出的核双子线性判别分析方法(KDS-DA),通过对LDA、DS-LDA、KDS-DA的层层深入研究,在双子空间中,将核方法与线性判别分析(LDA)方法相结合。通过matlab软件编程实现,证明这种方法具有更好的判别性能。 展开更多
关键词 人脸识别 线性判别分析 双子空间线性判别分析 双子空间线性判别分析
下载PDF
基于样本扩张和双子空间决策融合的单样本人脸识别算法 被引量:3
2
作者 杨军 袁红照 刘妍丽 《数据采集与处理》 CSCD 北大核心 2015年第1期148-154,共7页
对基于滑动窗口进行样本扩充的单样本人脸识别方法进行了改进,改进后算法一方面在识别阶段采用了比原算法更少的特征,提高了识别的时间效率;另一方面在训练阶段获得原始样本的镜像样本作为附加的训练、注册集合,通过学习训练形成双子空... 对基于滑动窗口进行样本扩充的单样本人脸识别方法进行了改进,改进后算法一方面在识别阶段采用了比原算法更少的特征,提高了识别的时间效率;另一方面在训练阶段获得原始样本的镜像样本作为附加的训练、注册集合,通过学习训练形成双子空间,识别结果由双子空间通过决策融合得到,提高了对测试样本变化的鲁棒性。在ORL人脸库和Feret子集人脸库上的实验表明,该算法在识别率上优于同类算法。 展开更多
关键词 人脸识别 单样本 样本扩充 滑动窗口 双子空间
下载PDF
求解大型稀疏线性系统的贪婪双子空间随机Kaczmarz方法 被引量:2
3
作者 荆燕飞 李彩霞 胡少亮 《同济大学学报(自然科学版)》 EI CAS CSCD 北大核心 2021年第10期1473-1483,共11页
基于一种有效的从系数矩阵中选取两个工作行的贪婪概率准则,提出一类求解大型稀疏线性系统的贪婪双子空间随机Kaczmarz方法。理论证明该方法收敛到相容线性系统的最小范数解,而且该方法的理论收敛因子小于原始双子空间随机Kaczmarz方法... 基于一种有效的从系数矩阵中选取两个工作行的贪婪概率准则,提出一类求解大型稀疏线性系统的贪婪双子空间随机Kaczmarz方法。理论证明该方法收敛到相容线性系统的最小范数解,而且该方法的理论收敛因子小于原始双子空间随机Kaczmarz方法的收敛因子。数值实验表明,该方法在求解性能方面较原始双子空间随机Kaczmarz方法更具优势。 展开更多
关键词 双子空间投影法 随机Kaczmarz方法 大型稀疏线性系统 收敛性分析
下载PDF
完备的双子空间边界近邻鉴别分析
4
作者 林玉娥 李敬兆 +1 位作者 梁兴柱 林玉荣 《计算机工程与科学》 CSCD 北大核心 2014年第7期1357-1362,共6页
针对边界Fisher鉴别分析算法不能够有效解决小样本问题,提出了一种完备的双子空间边界近邻鉴别分析算法。该算法通过理论分析将MFA的目标函数分解成两部分,对此目标函数的求解,首先要对高维样本进行PCA降维至一个低维子空间,而这一过程... 针对边界Fisher鉴别分析算法不能够有效解决小样本问题,提出了一种完备的双子空间边界近邻鉴别分析算法。该算法通过理论分析将MFA的目标函数分解成两部分,对此目标函数的求解,首先要对高维样本进行PCA降维至一个低维子空间,而这一过程并不损失任何有效的鉴别信息,对此通过定理1和定理2进行了证明;然后再分别求出类内边界近邻互补子空间的两投影矩阵。最后人脸库上的实验结果表明了所提方法的有效性。 展开更多
关键词 边界Fisher鉴别分析 小样本问题 双子空间 目标函数
下载PDF
双三相永磁同步发电机双子空间模型预测电流控制
5
作者 王玉雪 许德志 +2 位作者 赵文祥 郝亮 唐红雨 《电气工程学报》 CSCD 2022年第3期104-113,共10页
模型预测控制因其概念直观、响应迅速和便于处理非线性约束等优势,被广泛应用于多相电机控制系统。但现有的模型预测控制只考虑α-β基波子空间的机电能量转换,而未考虑因变流器控制死区和系统非线性因素引起的谐波损耗。针对双三相永... 模型预测控制因其概念直观、响应迅速和便于处理非线性约束等优势,被广泛应用于多相电机控制系统。但现有的模型预测控制只考虑α-β基波子空间的机电能量转换,而未考虑因变流器控制死区和系统非线性因素引起的谐波损耗。针对双三相永磁同步发电机整流系统,提出一种同时考虑α-β基波子空间和x-y谐波子空间以抑制谐波电流的双子空间多矢量模型预测电流控制策略。在此基础上,在两个子空间引入虚拟矢量占空比优化策略,从而进一步提升稳态控制性能。最后,通过试验验证了所提控制策略的可行性和有效性。 展开更多
关键词 双三相永磁同步发电机 谐波抑制 模型预测电流控制 双子空间控制 占空比
下载PDF
基于核的双子空间方法及其快速求解算法
6
作者 周晓彦 郑文明 +1 位作者 邹采荣 赵力 《模式识别与人工智能》 EI CSCD 北大核心 2010年第6期829-835,共7页
针对现有的双子空间方法中存在的问题,提出一种基于核的双子空间判别分析(KDS-DA)方法.此外,还提出一种基于镶边矩阵求逆运算的快速KDS-DA特征求解算法.该算法运用高阶镶边矩阵的求逆运算可转化为低阶镶边矩阵的求逆运算这一性质,使得... 针对现有的双子空间方法中存在的问题,提出一种基于核的双子空间判别分析(KDS-DA)方法.此外,还提出一种基于镶边矩阵求逆运算的快速KDS-DA特征求解算法.该算法运用高阶镶边矩阵的求逆运算可转化为低阶镶边矩阵的求逆运算这一性质,使得当顺序求解样本类内散射矩阵主空间中第r+1个KDS-DA判别矢量时,可充分利用求解第r个判别矢量时所得到的计算结果来减少算法复杂度.通过在ORL和AR人脸库上的实验证实文中方法的有效性. 展开更多
关键词 双子空间线性判别分析(DS-LDA) 核判别分析(KDA) 人脸识别
原文传递
A NOVEL APPROACH FOR DOA ESTIMATION IN UNKNOWN CORRELATED NOISE FIELDS 被引量:1
7
作者 Zhou Yi Feng Dazheng Liu Jianqiang 《Journal of Electronics(China)》 2006年第1期44-47,共4页
The key of the subspace-based Direction Of Arrival (DOA) estimation lies in the estimation of signal subspace with high quality. In the case of uncorrelated signals while the signals are temporally correlated, a novel... The key of the subspace-based Direction Of Arrival (DOA) estimation lies in the estimation of signal subspace with high quality. In the case of uncorrelated signals while the signals are temporally correlated, a novel approach for the estimation of DOA in unknown correlated noise fields is proposed in this paper. The approach is based on the biorthogonality between a matrix and its Moore-Penrose pseudo inverse, and made no assumption on the spatial covariance matrix of the noise. The approach exploits the structural information of a set of spatio-temporal correlation matrices, and it can give a robust and precise estimation of signal subspace, so a precise estimation of DOA is obtained. Its performances are confirmed by computer simulation results. 展开更多
关键词 Array signal processing Correlated noise Signal subspace Direction Of Arrival (DOA) estimation Spatio-temporal correlation matrices
下载PDF
New Two-Mode Coherent-Intermediate-Entangled State and Its Squeezing Operator
8
作者 XU Shi-Min XU Xing-Lei LI Hong-Qi 《Communications in Theoretical Physics》 SCIE CAS CSCD 2009年第11期808-812,共5页
The coherent-intermediate-entangled state |α, x)λ,v is proposed in the two-mode Fock Space, which exhibits both the properties of the coherent and entangled states. The |α, x)λ,v makes up a new quantum mechan... The coherent-intermediate-entangled state |α, x)λ,v is proposed in the two-mode Fock Space, which exhibits both the properties of the coherent and entangled states. The |α, x)λ,v makes up a new quantum mechanical representation, and the completeness relation of |α, x)λ,v is proved by virtue of the technique of integral within an ordered product of operators. The corresponding squeezing operators are derived. Furthermore, Generalized P-representation is constructed in the coherent-intermediate-entangled state |α, x)λ,v and the Schmidt decomposition of |α, x)λ,v is investigated. 展开更多
关键词 bipartite coherent-intermediate-entangled state IWOP technique classical transformation
下载PDF
求解相干线性方程组的稀疏惯性随机Kaczmarz算法
9
作者 梅娜 董巧丽 何松年 《数值计算与计算机应用》 2023年第4期392-408,共17页
双子空间投影算法和多步惯性随机Kaczmarz算法是求解相干线性方程组的有效算法,本文通过软阈值函数对这两种算法进行修正,提出了稀疏双子空间投影算法和稀疏多步惯性随机Kaczmarz算法,并给出其在有噪声干扰和无噪声干扰情况下在期望意... 双子空间投影算法和多步惯性随机Kaczmarz算法是求解相干线性方程组的有效算法,本文通过软阈值函数对这两种算法进行修正,提出了稀疏双子空间投影算法和稀疏多步惯性随机Kaczmarz算法,并给出其在有噪声干扰和无噪声干扰情况下在期望意义下的线性收敛率估计.通过数值实验验证本文所提算法的有效性和优越性. 展开更多
关键词 稀疏Kaczmarz算法 稀疏双子空间投影算法 稀疏多步惯性随机Kaczmarz算法 相干线性方程组
原文传递
Fisher线性鉴别函数的一种推广形式 被引量:3
10
作者 程正东 章毓晋 樊祥 《模式识别与人工智能》 EI CSCD 北大核心 2009年第2期176-181,共6页
鉴于常用的两种Fisher鉴别函数在应对奇异问题时存在的不足,给出一种Fisher鉴别函数推广形式,它能将双子空间鉴别分析中两个子空间的鉴别函数统一起来.本文还通过QR分解得到一个正交鉴别向量集,它与Fo-ley-Sammon正交鉴别向量集的鉴别... 鉴于常用的两种Fisher鉴别函数在应对奇异问题时存在的不足,给出一种Fisher鉴别函数推广形式,它能将双子空间鉴别分析中两个子空间的鉴别函数统一起来.本文还通过QR分解得到一个正交鉴别向量集,它与Fo-ley-Sammon正交鉴别向量集的鉴别性能很接近,但计算量较小.在2种人脸库上进行实验,实验结果与理论分析一致. 展开更多
关键词 Fisher鉴别函数 双子空间鉴别分析(DSDA) 广义特征值分解(GEVD) 正交鉴别向量集
原文传递
Bilocal Derivations of Reflexive Algebras
11
作者 荆武 王晓静 《Journal of Mathematical Research and Exposition》 CSCD 北大核心 2001年第2期197-202,共6页
Let A be a reflexive algebra in reflexive Banach space X such that both O+ ≠O and X_ ≠ X in LatA, then the set of all derivations of A into B(X) is topologically algebraically bireflexive.
关键词 reflexive algebra DERIVATION bifocal derivation.
下载PDF
THE REALIZATION OF MULTIPLIER HILBERT BIMODULE ON BIDUAL SPACE AND TIETZE EXTENSION THEOREM 被引量:1
12
作者 FANG XIAOCHUN(Department of Mathematics, Tongji University, Shanghai 200092, China) E-mail: xfang@online.sh.cn 《Chinese Annals of Mathematics,Series B》 SCIE CSCD 2000年第3期375-380,共6页
The multiplier bimodule of Hilbert bimodule is introduced in a way similar to [1], and its realization on a quotient of bidual space and Tietze extension theorem are obtained similar to that in C-algebra case. As a re... The multiplier bimodule of Hilbert bimodule is introduced in a way similar to [1], and its realization on a quotient of bidual space and Tietze extension theorem are obtained similar to that in C-algebra case. As a result, the multiplier bimodule here is also a Hilbert bimodule. 展开更多
关键词 Realization on bidual space Multiplier Hilbert bimodule Tietze extension theorem
原文传递
Two-photon reduction: a cost-effective method for fabrication of functional metallic nanostructures
13
作者 Sahar Tabrizi YaoYu Cao +1 位作者 Han Lin BaoHua Jia 《Science China(Physics,Mechanics & Astronomy)》 SCIE EI CAS CSCD 2017年第3期25-41,共17页
Metallic nanostructures have underpinned plasmonic-based advanced photonic devices in a broad range of research fields over the last decade including physics, engineering, material science and bioscience, The key to r... Metallic nanostructures have underpinned plasmonic-based advanced photonic devices in a broad range of research fields over the last decade including physics, engineering, material science and bioscience, The key to realizing functional plasmonie resonances that can manipulate light at the optical frequencies relies on the creation of conductive metallic structures at the nanoscale with low structural defects. Currently, most plasmonic nanostructures are fabricated either by electron beam lithography (EBL) or by focused ion beam (FIB) milling, which are expensive, complicated and time-consuming. In comparison, the direct laser writing (DLW) technique has demonstrated its high spatial resolution and cost-effectiveness in three-dimensional fabrication of micro/nanostrucmres. Furthermore, the recent breakthroughs in superresolution nanofabrication and parallel writing have significantly advanced the fabrication resolution and throughput of the DLW method and made it one of the promising future nanofabrication technologies with low-cost and scalability. In this review, we provide a comprehensive summary of the state-of-the-art DLW fabrication technology for nanometer scale metallic structures. The fabrication mechanisms, different material choices, fabrication capability, including resolution, conductivity and structure surface smoothness, as well as the characterization methods and achievable devices for different applications are presented. In particular, the development trends of the field and the perspectives for future opportunities and challenges are provided at the end of the review. It has been demonstrated that the quality of the metallic structures fabricated using the DLW method is excellent compared with other methods providing a new and enabling platform for functional nanophotonic device fabrication. 展开更多
关键词 two photon photoreduction metallic nanostructures nanofabrication PLASMONICS
原文传递
The Fourier space restricted Hartree-Fock method for the electronic structure calculation of linear poly(tetrafluoroethylene)
14
作者 FRIPIAT Joseph G. HARRIS Frank E. 《Science China Chemistry》 SCIE EI CAS 2014年第10期1355-1362,共8页
Building on the pioneering work of Jean-Marie Andre and working in the laboratory he founded, the authors have developed a code called FT-1D to make Hartree-Fock electronic structure computations for stereoregular pol... Building on the pioneering work of Jean-Marie Andre and working in the laboratory he founded, the authors have developed a code called FT-1D to make Hartree-Fock electronic structure computations for stereoregular polymers using Ewald-type con- vergence acceleration methods. That code also takes full advantage of all line-group symmetries to calculate only the minimal set of two-electron integrals and to optimize the computation of the Fock matrix. The present communication reports a bench- mark study of the FT-1D code using polytetrafluoroethylene (PTFE) as a test case. Our results not only confirm the algorith- mic correctness of the code through agreement with other studies where they are applicable, but also show that the use of con- vergence acceleration enables accurate results to be obtained in situations where other widely-used codes (e.g., PLH and Crys- tal) fail. It is also found that full attention to the line-group symmetry of the PTFE polymer leads to an increase of between one and two orders of magnitude in the speed of computation. The new code can therefore be viewed as extending the range of electronic-structure computations for stereoregular polymers beyond the present scope of the successful and valuable code Crystal. 展开更多
关键词 poly(tetrafluoroethylene) band structure Fourier-space Hartree-Fock lattice sum Ewaid
原文传递
Temporal behavior of low-amplitude two-photon screening-photovoltaic grey spatial solitons 被引量:1
15
作者 吉选芒 姜其畅 +1 位作者 王金来 刘劲松 《Optoelectronics Letters》 EI 2011年第3期225-228,共4页
The time-dependent formation of one-dimensional two-photon screening-photovoltaic(PV) grey spatial solitons under low-amplitude conditions is presented theoretically.The time-dependent propagation equation of two-phot... The time-dependent formation of one-dimensional two-photon screening-photovoltaic(PV) grey spatial solitons under low-amplitude conditions is presented theoretically.The time-dependent propagation equation of two-photon screeningphotovoltaic solitons is obtained by the numerical method.The results indicate that as the time evolves,the intensity width of grey screening-photovoltaic spatial solitons decreases monotonously to a minimum value towards the steady state.The higher the ratio of soliton peak intensity to the dark irradiation intensity,the narrower the width of grey solitons within the propagation time. 展开更多
关键词 Numerical methods PHOTONS SOLITONS
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部