xt3

刚体配准方法中，通常使用互信息函数作为相似性测度，此类提供计算的封装。(rigid registration method, which is commonly used as a mutual information function similar measure, the provision of such calculation package.)

对图像进行模糊处理，用逆滤波和维纳滤波恢复图像，也是实验时自己编写的代码，请多指教。(Fuzzy image processing, inverse filtering and Wiener filter to restore the image, but also experiments have written code, like him Zhijiao.)

CNN卷积神经网络，能以高速将图像精确到的分类，给力。(CNN convolutional neural network with high speed, accurate to classify images, awesome.)

基于光流场的交通汽车检测跟踪Traffic Vehicle Detection and Tracking Based on Optical Flow Field(Traffic Vehicle Detection and Tracking Based on Optical Flow Field)

hough变换提取椭圆,可以从二值图像中提取出椭圆(The ellipse can be extracted from Hough transform, and ellipse can be extracted from the two valued image.)

压缩感知，稀疏表示采用小波基表示，压缩测量采用随机高斯矩阵，重构算法是omp重构(Compressed sensing, sparse representation using wavelet representation, compression measurements using random Gaussian matrix remodeling reconstruction algorithm is omp)

本程序实现灰度图像各向异性扩散(Anisotropic Diffusion)滤波算法，可以用于平滑信号，保留边缘，增强边缘对比度。与普通高斯Gaussian算法相比，它不会平滑边缘的图像。(The program implements grayscale image using anisotropic diffusion filtering algorithm.It can be used to smooth the signal, keep the edge, enhance the edge of the contrast. Compared with ordinary Gaussian algorithm, it does not smooth the image edges.)

在matlab中读入视频序列，并显示出来。对视频流文件进行按帧读取，并计算帧与帧之间差值，并统计差值的分布。(Read into the video sequence in matlab, and displayed. Streaming video, read the press frame, and calculate the difference between frames, and statistical difference between the distribution.)

最小二乘匹配，图像匹配算法，需要初始匹配坐标，可以达到亚像素精度(least squares matching, image matching algorithm, the initial need to match coordinates, can reach sub-pixel accuracy)

基于visual c++平台，相机标定方法中的直接线性变换方法，即DLT方法(Based on visual c++ platform, camera calibration method of direct linear transformation method, namely DLT method)