In [ 3 ] , z . s . li , f . hall and c . eschenbash extended the concept of the base and period from nonnegative matrices to powerful sign patter matrices 文[ 3 ]中，李宗山等把非负矩阵的基和周期的概念推广到powerful符号矩阵。
Secondly , we utilize the nmf ( non - negative matrix factorization ) algorithm to extract human face local feature subspace 然后,对获得的类人脸肤色区域利用nmf ( non - negativematrixfactorization )非负矩阵分解的方法提取人脸局部特征子空间。
Some conditions are obtained by using the semigroup theory , the properties of nonnegative matrices and the techniques of inequalities to determine the asymptotically stable region of the equilibrium 通过半群理论、非负矩阵性质和不等式技巧，得到估计这类方程平衡态渐近稳定域的方法。
The holistic features are extracted by principal component analysis ( pca ) , and the local features are extracted by non - negative matrix factorization with sparseness constraints ( nmfs ) 首先通过主元分析算法( pca )提取全局特征,利用带稀疏限制的非负矩阵分解算法( nmfs )提取局部特征。
In this thesis , we mainly use snmf ( sparse nonnegative matrix factorization ) as the method of rank reduction , which extend the nmf to include the option to control sparseness explicitly 本文主要采用snmf (非负稀疏矩阵分解)算法作为降维和提取特征向量的工具,该算法是在nmf算法的基础上加上显式地稀疏因子控制而形成的一种非负矩阵分解方法。
Principle component analysis ( pca ) , as a classical method for feature extraction , learns holistic representations of facial images , while non - negative matrix factorization ( nmf ) , a recently proposed approach , learns parts - based representations of faces . however , we argue that nmf can not only learn parts - based representations but also holistic ones with different sparseness constraints 在众多的特征提取算法中,基于全局特征提取的主元成分分析( principlecomponentanalysis , pca )是讨论最多的经典算法,与此对应的是基于局部特征提取的非负矩阵分解( non - negativematrixfactorization , nmf )算法。
The explicit method is widely used for its simpleness and little memory consumed with local time step and variable coefficients implicit residual smooth to accelerate the convergence procedure . according to yoon and jameson ' s ideas , an efficient implicit lu - sgs algorithm is carefully constructed by combing the advantages of lu factorization and symmetric - gauss - seidel technique in such a way to make use the l and u operators scalar diagonal matrices , thus the numeric algorithm requires only scalar inversion . the computational efficiency is greatly improved with this scheme 显式方法具有简单，消耗内存小等优点，并采用当地时间步长、变系数隐式残值光顺等加速收敛措施，在定常流动的模拟中得到了广泛的应用;根据yoon和jameson提出的简化正、负矩阵分裂，构造的l 、 u算子只需进行标量对角阵求逆，极大提高了流场数值求解过程的计算效率;采用newton类型的伪时间子迭代技术使时间推进精度提高至二阶。
In this thesis , we propose an efficient nmfs + rbf aggregate framework for fr , in which non - negative matrix factorization with sparseness constraints ( nmfs ) is firstly applied to learn either the holistic representations or the parts - based ones by constraining the sparseness of the basis images , and then the rbf classifier is adopted for pattern classification 本文提出了一种基于非负矩阵稀疏分解( non - negativematrixfactorizationwithsparsenessconstraints , nmfs )和rbf神经网络的人脸识别方法。通过控制稀疏度, nmfs算法既可提取人脸全局也能提取局部特征,再运用rbf神经网络进行模式分类。
Different from other rank reduction methods , such as pca ( principal component analysis ) and vq ( vector quantization ) , nmf ( nonnegative matrix factorization ) can get nonnegative , sparse basis vectors which make possible of the concept of a parts - based representation 与pca (主分量分析)和vq (矢量量化)等降维算法不同, nmf (非负矩阵分解)算法能够分解出非负的,稀疏的特征矩阵和编码矩阵,能够提取原始数据向量的局部特征,使基于局部特征进行分类的聚类算法更容易实现。
