VS2019支持包.rar
MATLAB对应的VS2019支持包
msvc2019.xml与msvcpp2019.xml
image quilting.rar
image quilting for texture synthesis and transfer:图像缝合的纹理合成和转移
k-means.rar
聚类算法有很多种,K-Means 是聚类算法中的最常用的一种,算法最大的特点是简单,好理解,运算速度快,但是只能应用于连续型的数据,并且一定要在聚类前需要手工指定要分成几类。
DBSCAN.rar
DBSCAN(Density-Based Spatial Clustering of Applications with Noise,具有噪声的基于密度的聚类方法)是一种很典型的密度聚类算法,和K-Means,BIRCH这些一般只适用于凸样本集的聚类相比,DBSCAN既可以适用于凸样本集,也可以适用于非凸样本集。下面我们就对DBSCAN算法的原理做一个总结。
密度聚类原理
DBSCAN是一种基于密度的聚类算法,这类密度聚类算法一般假定类别可以通过样本分布的紧密程度决定。同一类别的样本,他们之间的紧密相连的,也就是说,在该类别任意样本周围不远处一定有同类别的样本存在。
able-running.rar
有限元方法入门:有限元方法简单的一维算例。
算例描述
我们对下述边值问题\label{eq1}
{u′′(x)+u(x)=(1−π2)sinπxu(0)=u(1)=00≤x≤1
{u″(x)+u(x)=(1−π2)sinπx0≤x≤1u(0)=u(1)=0
考虑其变分问题,对其变分问题有限元离散并求解,并验证其收敛阶。
注:问题的真解为u(x)=sinπxu(x)=sinπx。
CUDA Toolkit Reference Manual
CUDA_Toolkit_Reference_Manual CUDA开发者必备的API参考手册
CUDA Runtime API
NVIDIA CUDA Runtime API
API Reference Manual
The driver and runtime APIs are very similar and can for the most part be used
interchangeably. However, there are some key differences worth noting between the
two.
数字图像处理 国外教材
数字图像处理
国外系列教材
数字图像处理
《国外电子与通信教材系列·数字图像处理(第2版)》是数字图像信息处理领域的一本经典著作,是20多年来此领域最权威的教材之一。与1977年问世的《国外电子与通信教材系列·数字图像处理(第2版)》第一版相比,进行了重要修订和扩充,增加了关于小波变换、图像形态学和彩色图像处理的章节,并新增了500多幅图像和200多幅图表。全书共分12章,首先介绍了数字图像基础、空间域和频域的图像增强;然后讨论了图像复原、彩色图像处理、小波变换及多分辨率处理和图像压缩;最后讲述了形态学图像处理、图像分割、表示与描述以及目标识别等。
BRISQUE算法的MSCN系数分布图
BRISQUE算法的MSCN系数分布图 matlab 代码
Neural Networks and Deep Learning
Neural Networks and Deep Learning
Make Your Own Neural Network
用很浅显的文字介绍了神经网路及BP算法,适合机器学习入门
kinect_emgu_图像格式转换
ImageExtemsion.cs:提供一些静态方法,协助你可以把Kinect SDK的深度影像与彩色影像转换成WPF的Bitmap或BitmapSource类别储存影像信息。
EmguImageExtension.cs:里面会用到ImageExtemsion.cs档案中的方法,把KinectSDK的彩色影像转换成EmguCV的格式。
kurtosis.m
k = kurtosis(X) returns the sample kurtosis of X. For vectors, kurtosis(x) is the kurtosis of the elements in the vector x. For matrices kurtosis(X) returns the sample kurtosis for each column of X. For N-dimensional arrays, kurtosis operates along the first nonsingleton dimension of X.
k = kurtosis(X,flag) specifies whether to correct for bias (flag is 0) or not (flag is 1, the default). When X represents a sample from a population, the kurtosis of X is biased, that is, it will tend to differ from the population kurtosis by a systematic amount that depends on the size of the sample. You can set flag to 0 to correct for this systematic bias.
k = kurtosis(X,flag,dim) takes the kurtosis along dimension dim of X.
kurtosis treats NaNs as missing values and removes them.
BLLIINDS_II
we develop an efficient BLIINDS-II algorithm using NSS approach in DCT domain for image and video quality assessment with no reference image. The approach relies on a simple Bayesian inference model to predict image and video quality score, after a set of features is extracted from an image. These features are extracted from a generalized NSS based model of local DCT coefficients. Generalized Gaussian density model parameters are used to form these features. BLIINDS-II (Blind Image Integrity notator using DCT Statistics-II) adopts a simple probabilistic model for score prediction. Given the extracted features from a test image/video, the quality score that maximizes the probability of the empirically determined inference model is chosen as the predicted quality score of that image/video. When tested on the LIVE IQA database, BLIINDS-II correlates highly with the human judgments of quality.
BLIINDS算法
由于自然图像具有高度结构化特性,人类视觉特性(Human Visual System,HVS)倾向于从视觉图像中提取结构信息,并且对图像的对比度高度敏感。Saad等
人提出了一种基于DCT域统计特性的图像完整性评价方法(BLIINDS)。BLIINDS提取 DCT 系数的统计特征,利用机器学习的方法来评价图像质量。