搜索资源列表
wine_Gauss-
- 采用高斯核的多类分类,以wine数据位例，进行编程-the classification using gauss kernal
mpmr_code
- 最小最大概率机分类器，包括线性核函数和高斯核函数两种-Minimax probability machine classifier
EM_Algorithm
- 介绍期望最大算法基本原理及聚类实现，可以很好的对多个高斯概率密度分布进行分类-Introduces the basic principle and expectation maximization clustering algorithm to achieve, can be good for multiple Gaussian probability density distribution of the classification
classifier
- 简单的分类小程序。包括高斯混合模型、svm（调用函数）、logistic regression、和人工神经网络-Simple classification applet. Including the Gaussian mixture model, svm (calling function), logistic regression, and artificial neural networks
fenlei
- 空间约束半监督高斯过程下的高光谱图像分类 -Space constraints Semi-supervised hyperspectral image classification under the Gaussian process
ICA
- ICA快速算法，提高分类效果 滤波 源信号所含的高斯白噪声越多，分离后得到的信号与源信号相比误差越大，效果越差；所含高斯白噪声越少，分离效果越好-Fast ICA algorithm to improve the classification results more contained in the source signal filtering Gaussian white noise, the signal obtained with the source signal separatio
EM
- EM 分类 多元高斯分布的一段程序 分类效果比较不错 -A program classification results EM classification multivariate Gaussian distribution is quite good
dd_ex6
- 高斯分布混合监督之间的区别，训练三个分类器。检测异常。-The distinction between mixed Gaussian distribution supervision, training three classifiers.
gmm
- 混合高斯模型使用K（基本为3到5个） 个高斯模型来表征图像中各个像素点的特征,在新一帧图像获得后更新混合高斯模型,用当前图像中的每个像素点与混合高斯模型匹配,如果成功则判定该点为背景点, 否则为前景点。通观整个高斯模型，他主要是有方差和均值两个参数决定，,对均值和方差的学习,采取不同的学习机制,将直接影响到模型的稳定性、精确性和收敛性。由于我们是对运动目标的背景提取建模，因此需要对高斯模型中方差和均值两个参数实时更新。为提高模型的学习能力,改进方法对均值和方差的更新采用不同的学习率 为提高在繁忙
Kmeans_Gauss_Bayes
- 用matlab实现高斯混合模型的前期处理和分类训练 - Using Gaussian mixture model matlab realize the pre-treatment and classification of training!
EMal
- 使用EM算法完成图像的分割，使用混合高斯模型，可以修改分类数。运行EM.m-Expectation Algorithm
speech-emotion-recognition
- 过特定人语音情感数据库的建立；语音情感特征提取；语音情感分类器的设计，完成了一个特定人语音情感识别的初步系统。对于单个特定人，可以识别平静、悲伤、愤怒、惊讶、高兴5种情感，除愤怒和高兴之间混淆程度相对较大之外，各类之间区分特性良好，平均分类正确率为93.7 。对于三个特定人组成的特定人群，可以识别平静、愤怒、悲伤3种情感，各类之间区分特性良好，平均分类正确率为94.4 。其中分类器采用混合高斯分布模型。-The system of speech emotion recognition
autogaussianGaborfits
- 主要实现了高斯-gabor小波的匹配,两者的结合使其在分类的效率得到提高，适用于图像处理的分类和识别-The main achievement of matching Gauss-gabor wavelet, so the combination of the two categories of increased efficiency, suitable for classification and identification of image processing
project
- 语音增强，加权去噪自动编码器，信噪比估计，维纳滤波，噪声分类，高斯混合模型,维纳滤波-Wiener filtering based speech enhancement
gaosihefenlei
- 用于SVM高斯核分类-Gaussian kernel for SVM classification-Gaussian kernel for SVM classification
Bayes
- 朴素贝叶斯分类器，能实现高准确的分类，且速度快-Naive Bayes classifier, can achieve high accuracy of classification and fast
Neighbor-classification
- 近邻分类，属于模式识别类的。对两类数据分别产生高斯分布数据，用KNN算法看这个数属于哪个类的，并求测试数据和类中每个数据的欧氏距离-Neighbor classification, pattern recognition belongs to the class. Two types of data were generated Gaussian distribution data, to see with this number belongs KNN algorithm class and
guassian-discriminant-analysis
- 基于opencv实现的c++版高斯判别分类，随机生成两类满足正态分布的训练样本点，利用高斯判别分析进行分类-Based on c++ version of the opencv, we realized gaussian discriminant classification by randomly generating two classes meet normal distribution of the training sample points, and used gaussian di
gpc_prepare
- 高斯过程二分类的预处理函数，它估计了log(q(y|X,θ-gaussian processes classification prepare function
approx_for_gpc
- 高斯过程二分类中，后验概率的几种估计算法-Approximations for Binary Gaussian Process Classifi cation