搜索资源列表
图像纹理分析-节省时间
- 图像纹理分析的好的源代码，可以计算纹理的二阶特征：能量，熵，局部稳定性，惯性距，和相关性。节省了分析纹理的研究人员的时间。-image texture analysis of good source code can calculate the second-order texture characteristics : energy, entropy, local stability, inertial distance, and relevance. Texture analysis of
diagnosis
- 小波包特征熵的飞机液压系统故障诊断diagnosis方法-Entropy of wavelet packet features of the aircraft hydraulic system fault diagnosis
Texture-feature-code-matlab
- matlab最全纹理特征代码，灰度共生矩阵软件自带，就不传了。里面有tamura六个参数，灰度梯度共生矩阵的纹理特征，还有熵。-matlab the best texture feature code, the GLCM software comes not passed. Tamura six parameters, the shades of gray co-occurrence matrix texture features, as well as entropy.
Endpoint-detection
- 基于短时能量与过零率、倒谱特征和谱熵的三种语音端点检测-Endpoint detection
med2d
- 一种基于最小熵反褶积( ]8+83)3 6+D*(A4 Q20(+W(-)D8(+，]6Q) 的滚动轴承故障特征提取方法: 在利用 /X 模型去除齿轮啮合产生的确定性信号的基础上，对保留信号进行最小熵反褶积，增强冲击信号"该方法避免了传统轴承故障诊断方法中带通滤波器设计的难题，实车测试表明: 与共振解调技术相比，该方法提取的滚动轴承故障特征更加明显，更适合于工程应用"-Based on minimum entropy deconvolution (] 8+83) 3 6+D* (A4 Q20 (
SE
- MATLAB小波熵程序，用于对信号进行分析，可较容易得到信号特征。-MATLAB wavelet entropy procedure, used to signal analysis, can more easily get the signal characteristics.
glcm
- 基于灰度共生矩阵的图像特征提取,包括熵、相关性、标准差及方差-Image feature extraction based on GLCM，including entropy, correlation, standard deviation and variance
GLMCandLBPextraction
- 第一个程序提取了图像灰度级为64的灰度共生矩阵，并计算了能量，熵，对比度，相关性，逆差矩这5个参数.第二个程序可以提取彩色图像的LBP纹理特征，可以提取采样点为8、16、24的统一模式（u2）、旋转不变模式（ri）、统一旋转不变模式（riu2）的LBP值。-The first program to extract a grayscale image GLCM 64, and calculate the energy, entropy, contrast, correlation, inverse
Renyi
- 计算一维时间序列的Renyi熵，可作为脑电信号的特征提取方法，从而对脑电的复杂度进行分析-The Renyi entropy of one dimensional time series can be calculated as a feature extraction method of EEG signal, which can be used to analyze the complexity of EEG.
HHT
- 把HHT和神经网络结合起来，识别出是否故障。对于HHT变换，这里用到的是它的IMF分解，然后用能量理论来判别哪些模组是虚假分量，哪些是是真实分量。对于EMD分解后的谱进行特征提取 ，利用的理论基础就是模糊熵。计算出真实 分量的模糊熵，作为输入层；输出层为两个神经元，为0（故障 ）1（正常 （正常 ）判别该信号是否出现故障。-The combination of HHT and neural network to identify whether the fault. For the HHT tr
小波分解+MSE熵+特征提取
- 通过小波四层分解，获取脑电信号的四个波形，之后运用多变量多尺度熵计算综合MMSE，为脑电信号识别做基础(After four layers of wavelet decomposition, four waveforms of EEG signals are acquired, and then the MMSE is calculated by using multivariable and multi-scale entropy, which is the basis of EEG reco
SampEn
- 模糊熵特征提取 计算信号的复杂度 用于信号识别 故障诊断(Fault diagnosis for signal recognition)
分形维数和近似熵
- 分形维数和近似熵 用于提取特征量 实现信号识别 故障诊断(Fault diagnosis based on signal recognition)
图像熵
- 图像熵在图像分割、图像识别领域应用广泛，如果考虑各像素之间统计独立，图像的一维熵可以表示图像灰度分布的聚集特征。(Image entropy is widely used in image segmentation and image recognition. If the statistical independence between pixels is considered, the one-dimensional entropy of the image can represent th
REC-FSA-master
- 利用信息熵聚类对故障多特征量进行特征选择(Feature selection by using information entropy clustering for multiple features of fault)
模糊熵程序
- 本文件为一个轴承故障诊断指标，名字为模糊熵，可以作为特征量的筛选指标，也可用作为特征指标。(This document is a bearing fault diagnosis index, whose name is fuzzy entropy, which can be used as a screening index of characteristic quantity, and can also be used as a characteristic index.)
信息论特征选择KDD Code
- 基于信息熵的特征选择算法，评价每个属性与分类的关联信息，评价属性，进行特征选择(Feature selection algorithm based on Information Entropy)
程序
- 脑电信号特征提取 熵算法 信号处理 非线性分析方法 近似熵 样本熵 排列熵(Feature Extraction of EEG Signals Entropy algorithm signal processing Nonlinear Analysis Method Approximate Entropy Sample Entropy Arrangement Entropy)
程序
- main_feature.m：为特征提取主要程序,其中调用filter50.m子程序为50HZ工频滤噪；调用ApEn.m c0complex.m kEn_correct.m lyapunov_wolf.m LZC.m spectral_entropy.m SVDen.m SampEn.m子程序为非线性特征（近似熵，C0复杂度，K熵等）提取；wave_brain为小波分析频段特征提取。其中采样频率皆为256HZ。(ApEn.m c0complex.m kEn_correct.m lyapunov_
CEEMD-样本熵
- 用ceemd分解信号IMF分量，用峭度相关原则筛选噪声，用样本熵进行特征提取(18/5000 CEEMD was used to decompose the signal and sample entropy was used to extract the feature)