搜索资源列表
R-KDDA
- 核直接线性判别方法:图像及高维复杂数据模式识别的利器!内有方法开发的相关文档说明!经典!-The matlab functions implement the methods presented in the paper [TNN_KDDA02.pdf] Juwei Lu, K.N. Plataniotis, and A.N. Venetsanopoulos, "Face Recognition Using Kernel Direct Discriminant Anal
OSU_SVM3.00
- 支持向量机可用于数据约简以及分类,它是将原始空间的样本通过非线性映射变换到高维特征空间的方法-Support Vector Machines can be used for data reduction and classification, it is the original sample space through nonlinear mapping transformation to high-dimensional feature space method
595643603713295726
- kfcm,为模糊核聚类算法,用于将低维的数据映射到高维进行分类,是较先进的算法-kfcm, the fuzzy kernel clustering algorithm for low-dimensional data is mapped to high-dimensional classification, is a more advanced algorithms
DrawGuassin
- matlab代码,描述了如何画高斯一维分布,和二维,三维以及高维分布,有产生高斯数据的matalb函数-matlab code,discr ipt how to draw a gaussian function and the code that produce the data for this fuction
cede
- 计算测地距离,测地距离是众多距离的一种,在高维流行数据分布空间上,欧式距离不再适用,这时就可用到欧氏距离-Computing geodesic distance, geodesic distance is a large distance, popular in high-dimensional data distribution space, the Euclidean distance is no longer applicable, then you can use the Euclide
manifolds
- 流形学习是近年来机器学习及模式识别等领域的一个研究热点,其主要目标是去发现高维观察数据空间的低维光滑流形。自从2000年Roweis和Saul提出LLE算法、Tenenbaum等人提出Isomap算法,特别是Donoho等人发现Isomap算法能够准确发现人脸图像流形潜在的参数空间、张长水等人将LLE算法用于人脸识别并取得了较好的识别效果之后,基于流形学习的人脸识别研究引起了人们的广泛关注。-Manifold learning in recent years the field of machi
PCA
- PCA算法主要用于提取高维特征,可以将图像这种高位数据降维成为低维向量-PCA algorithm
mddm(data)
- MDDM多标记降维,可以用来减少高维多标记数据的维度。-MDDM is a package for multi-label dimensionality reduction. It can be used to reduce the dimensionality of high-dimensional multi-label data.
duoyuantongji
- 多元统计方法是处理高维、相关数据的有效工具,其首先 在化工过程的故障检测中得到成功应用,后也被应用于其他的 工业过程-Multivariate statistical methods to deal with high dimension and effective tool for data, the first fault detection in chemical process has been successfully applied, the latter also used
treelet-testcode_April08
- 一维无序随机信号的小树变换处理,小树变换即treelet,是一种有效的高维的,无序的,杂乱的数据进行降维,特征提取的方法-One-dimensional disorder random signal processing, the saplings transform treelet tree transform namely, is a kind of effective high-dimension, disorder, messy data dimension reduction, fea
matlab_shibie
- 基于神经网络的图像识别的基本原理。基于图像像素数据的神经网络识别技术,是用高维的原始图像数据作为神经网络的训练样本。-Based on neural network image recognition of the basic principles. Image pixel data based on neural network identification technology is the use of high-dimensional raw image data as a neural
lle
- 局部线性降维方法LLE 用于高维空间数据的降维处理-Local linear dimension reduction methods for high dimensional data LLE dimensionality reduction process
lle
- 流形学习,局部线性嵌入式算法(LLE),一种智能的算法去推测捕捉高维空间中所包含的低维特征。与适合于局部维数约减的聚类算法不同,LLE算法在单一的低维的全域坐标系统中表征采样空间,然而它并没有优化最小局域。通过对线性重构的局域对称的研究应用,LLE能够描述非线性流形的全局结构,例如那些人脸的数据集或者文本文档集-Manifold learning, embedded local linear algorithm (LLE), an intelligent algorithm to predict
kpca1
- KPCA 的基本思想是将数据从输入空间映射到高维特征空间,然后在特征空间利用线性主成分分析方法计算主成分。本程序是KPCA的源程序,可实现调用。对于初学者或许有帮助。-The basic idea of KPCA is to map the data from the input space to high dimensional feature space, then in the feature space using principal component analysis method
ComSam
- Compressive Sampling是最近讨论非常热烈的数据处理方法。用一个与变换基不相关的观测矩阵将变换所得到的高维信号投影到一个低维空间上,然后通过优化求解从这些少量的投影中以高概率重构出原信号。-Compressive Sampling is a very lively discussion of recent data processing method. A base with the transformation matrix will not change the releva
LLE
- 高维非线性数据维数约减,局部线性嵌入算法-Dimension reduction of the higher dimensional nonlinear data,locally linear embedding.
KPCA
- KPCA是一种基于核的主要成分分析,是一种由线性到非线性之间的桥梁。通过非线性函数把输入空间映射到高维空间,在特征空间中间型数据处理,引入核函数,把非线性变换后的特征空间内积运算转换为原始空间的核函数计算。 基本思想是通过某种隐士方法将输入空间映射到某个高维空间(特征空间),并在特征空间实现PCA。对该算法进行了详细的说明-KPCA is a kernel-based principal components analysis, is a bridge between the linear
K_l
- 基于K-L变换的特征值提取 三维 数据效果还可以高维转换为低维-K-L character
2D-LDA
- LDA是一种线性降维方法,对原有的高维人脸数据集降维,然后识别,具有很好的聚类和识别效果。有详细的说明-LDA is a linear dimensionality reduction method, the original high-dimensional face data set dimensionality reduction, and then identify clustering and identification. Described in detail
KDTreeTest
- 这是一个名为kdtree的数据结构源码,用来划分高维空间,并且查找最近的点-this source code is a kind of structure to split the space of multi-dimension, which is named kdtree. And you can use it to compute the nearest point in the database.