Mumford

提供了两种光照补偿的方法进行比较，一种是参考白，一种是非线性补偿，并添加了直方图比较和参考文档。(Offers two illumination compensation methods were compared, one is the reference white, one is non-linear compensation, and adds a histogram comparison and reference documentation.)

利用fastICA算法实现了两幅图像的分离。内附源图像文件（两种）。同时还利用峰值信噪比PSNR和性能指标PI实现了对分离效果的评价。原创代码。(fastICA algorithm is uesd in seperating the mixed images.)

说明：  基于高斯混合模型和隐马尔科夫模型的图像分割算法(Image segmentation algorithm based on Gaussian mixture model and hidden Markov model)

基于偏微分方程的遥感图像处理方法 利用测试图像对算法进行仿真验证，并与其它图像去噪算法进行去噪效果比较，通过数值对降噪效果进行评估，得出该方法不仅能够去除噪声而且能够很好的保护图像的边缘等细节信息，弥补了高阶非线性扩散去噪方法导致边界模糊的缺陷，提高了遥感图像的质量，改善了视觉效果(Simulation algorithm of remote sensing image processing methods based on partial differential equations using the test image denoising results compare with other image denoising algorithm to assess the value of the noise reduction effect, draw the method can not only remove the noise but well protected image edge details, to make up for the higher order nonlinear diffusion denoising method lead to fuzzy boundaries defects and improve the quality of remote sensing images and improve the visual effects)

在对散乱点云进行三维重建时，许多算法都需要求点云的法矢并保证所有法矢的方向一致向外。本程序就是对点云数据求法矢和方向调整，包含文件输入输出的完整程序。(Scattered normals strike and reorientation)

利用python和opencv拼接两张图片(Splicing two pictures using Python and opencv)

说明：  小波阈值去噪。对灰度图像和二值图像进行滤波。(Wavelet thresholding filter . Denoiseing for the gray image and binary image .)

利用MATLAB实现了视频图像行人识别与检测，(Using MATLAB to achieve video image pedestrian recognition and detection,)

可以在harris角点检测和ncc粗匹配之后实现精确准确的角点匹配，为下一步配准做准备(Can the Harris corner detection and NCC coarse matching after achieve precise accurate angular point matching, preparing for the next registration )

说明：  matlab生成的fig图纸转换为DXF文件(Converting fig drawings generated by MATLAB into DXF files)