几种经典的图像噪声方差估计方法

快速运动去模糊，在盲去卷积过程中运用一些加速算法(Fast motion to blur in the blind deconvolution algorithm to accelerate the process of using some)

基于相位一致性与K-means方法的结合边缘检测(PC and K-means)

西南交通大学出版社出版的《信号与图像的稀疏分解及初步应用》，作者是王建英老师，尹忠科老师和张春梅老师(Southwest Jiaotong University Press sparse signal and image decomposition and preliminary application, WANG Jian, teacher, Yin Zhongke teacher and Zhang Chunmei teacher)

matlab结合GPU的资料，初学者可以看看，很不错的解释,适合参考(matlab GPU combined data, beginners can see, very good explanation for the reference)

基于粒子滤波的视频目标跟踪算法matlab源程序，内涵视频文件及算法源程序。(Video object tracking algorithm based on particle filter matlab source program, content video files and algorithm source program.)

基于opencv实时图像的2值化处理，一定要有摄像头(Opencv-based real-time image processing of two values, there must be a camera)

8*8矩阵dct变换以及量化 反量化和反dct变换(8* 8 transform and quantization matrix dct quantization and anti-anti-dct transform)

基于Mean Shift的图像分割过程就是首先利用Mean Shift算法对图像中的像素进行聚类，即把收敛到同一点的起始点归为一类，然后把这一类的标号赋给这些起始点，同时把包含像素点太少的类去掉。然后，采用阈值化分割的方法对图像进行二值化处理 (Mean Shift Based on the process of image segmentation is the first to use the image Mean Shift algorithm for clustering of pixels, that is, to converge to the same point as a starting point to the category and then assigned to this type of labeling of these starting points, at the same time to contain too few pixels to remove the category. Then, using thresholding segmentation method of binary image processing)

人脸识别源代码，分为人脸检测，人脸识别等几部分(Face recognition source code, is divided into face detection, face recognition and other parts)

基于VC++的图像处理类，可以用于bmp文件的读写，显示(Based on the image processing VC++ classes that can be used to read and write bmp files, display)