criminisi-source-code

一种基于SIFT的图像配准方法，目前使用最广泛的一种配准方法。性能优。(One kind SIFT image registration method based on the most widely used kind of registration method. Excellent performance.)

程序将界面与数据处理分开，界面采用MFC的对话框模板，对注册、训练、识别和主界面分别建立一个类，通过消息处理函数与用户进行交互，利用多线程来实时显示图像。 数据处理又分为了CFaceAlign（人脸检测+几何归一化）、CLightPrep（光照归一化）、CFaceFeature（Gabor特征提取）、CSubspace（计算Fisherface子空间）四个类，还有一个类 CFaceMngr 负责管理界面与数据之间的交流。这样的程序结构使对界面和算法彼此独立，互不影响。 每个类的具体方法请见相应头文件中的注释。(The program will interface with the data processing separate interface using MFC dialog template, registration, training, recognition and the main interface, respectively, to create a class that interacts with the user through a message handler, using multiple threads to display images in real time.   Data processing is divided into a CFaceAlign (Face Detection+ geometric normalization), CLightPrep (illumination normalization), CFaceFeature (Gabor feature extraction), CSubspace (computing Fisherface subspace) four categories, as well as a class is responsible for managing CFaceMngr interface and data exchange between. Such a program structure allows for interfaces and algorithms are independent of each other. Specific methods of each class, see the comments in the appropriate header files.)

基于sift算法的图像拼接matlab代码，经过测试，可以运行，对有重叠部分的图像进行拼接，效果不错(Sift algorithm based on the image mosaic matlab code, the test run splicing with overlapping portion of the image, good results)

正余弦+NLM滤波器和正余弦与其他滤波器的视觉效果与数值信息比较(Sine and Cosine+NLM cosine filter and compare numeric information with visual effects and other filters)

基于MATLAB的SOM神经网络设计源程序(MATLAB-based SOM neural network design source)

本书是计算机视觉编程的权威实践指南,依赖 Python 语言讲解了基础理论与算法,并通过 大量示例细致分析了对象识别、基于内容的图像搜索、光学字符识别、光流法、跟踪、三维重建、 立体成像、增强现实、姿态估计、全景创建、图像分割、降噪、图像分组等技术。(This book is an authoritative guide to computer vision programming, which explains basic theories and algorithms based on the Python language.A large number of examples carefully analyze object recognition, content based image search, optical character recognition, optical flow, tracking, 3D reconstruction,Stereoscopic imaging, augmented reality, attitude estimation, panoramic creation, image segmentation, noise reduction, image grouping and other techniques.)

锐化效果显示图像 例子

说明：  meanshift算法实现目标的跟踪，由视频序列文件中提取出第一帧，定义目标区域，进行跟踪，动态显示跟踪结果(meanshift target tracking algorithm, by the video sequence documents to extract the first frame, the definition of the target area, tracking, tracking the results of dynamic display)

说明：  ：建立了碳钢渗硫表面和非渗硫表面的多球凸峰面接触模型，模型中，凸峰高度按高斯函数分布．用塑性流线 理论分析了渗硫层的摩擦系数．计算结果表明：摩擦系数随载荷的增加而增加；相同粗糙度下，磨削加工的摩擦系 数比车削加工的摩擦系数大；摩擦系数随着渗层厚度的增大而增大．从减小摩擦系数的角度来考虑，渗硫层并非 越厚越好． (the multi-asperity sphere-shape model for surface contact between sulphurized layer and unsulphurized layer is established．in which the height of sphere-shape asperities follows the Gaussian distribution．1he frictional coeficients of sulphurized layers are analyzed by using plastic flown line theory．1he results show that the frictional coeficient increases with the load．the frictional coeficient of ground surface is higher than that of tumed surface in case of the same roughness．1he thicker the sulphuri[ed layer is．the higher the frictional coeficient is．From the viewpoint of reducing frictional coeficient，the thickness of sulphurized layer should be smaller．)

稀疏贝叶斯学习的代码，出自UCLA，大家可以下载看看(sparse bayesian learning)