GMMP-and-tracking

本程序将SIFT与边缘结合起来用来做图像配准 先用SIFT检测关键点 用CANNY算子进行边缘检测 进而优化选取最佳匹配点 进行向量匹配 图像配准。 (The program will combine the SIFT and the edge of image registration is used to make the first detection of the key points with SIFT operator with CANNY edge detection and then select the best match point optimized vector matching image registration.)

基于matlab开发的因离焦产生的图像模糊处理，利用盲卷积处理方法，取得很好的效果(blind deconvolation)

自己编的改进谐波小波包实现程序，实现了信号经过谐波小波包变换后幅值不变，适合于机械故障特征提取(The own series improved harmonic wavelet packet achieve signal amplitude unchanged after harmonic wavelet packet transform, for mechanical fault feature extraction)

本程序是用似然比检验的方法进行profile中的变点检测(This procedure is used likelihood ratio test method of change point detection profile)

鼠标滑动截图，截图后 点击winform其它位置 则将图片黏贴到相应位置

mser算法（Maximally Stable Extremal Regions）代码，非文章作者写的，但是还不错，matlab实现。(mser algorithm (Maximally Stable Extremal Regions) code, non-article the authors write, but also good, matlab realize.)

说明：  图像分割的MATLAB实现，效果很好，对光照有一定的适应性(Image Segmentation of MATLAB to achieve good effect on the light a certain degree of adaptability)

建立了普通话语音性别数据库,提出联合梅尔频率频谱系数(Mel2f requency Cep st rum Coefficient s , MFCC) 的特征提取方法和支持向量机(Support Vector Machine , SVM) 的分类方法进行说话人性别识别,并与其它分类方法进行比较。(A Chinese speech ( mandarin ) database was established for speaker s gender recognition. A combination met hod is p roposed for gender recognition of speaker s based on support vector machine and Mel2f requency cep st rum coefficient s (MFCC) for classification and feat ure ext raction respectively.)

基于matlab的高斯金字塔算法，和一般的算法不同，此算法非常详细，逐步编写(Gaussian pyramid matlab-based algorithm, and the different algorithms, this algorithm is very detailed, step by step to prepare)

基于SIFT的图像配准程序 SIFT特征匹配算法是目前国内外特征点匹配研究领域的热点与难点,其匹配能力较强,可以处理两幅图像之间发生平移、旋转、仿射变换情况下的匹配问题,甚至在某种程度上对任意角度拍摄的图像也具备较为稳定的特征匹配能力(Hot and difficult of the current research areas of domestic and foreign feature point matching based the SIFT image registration program SIFT feature matching algorithm is its ability to match between the two images, can handle translation, rotation, affine transformation case match problem, even in some degree images taken at any angle with more stable feature matching ability)