VisualSFM_windows_cuda_64bit

根据RBM深度网络（Hinton，2006）进行MINST手写文字的分类器的训练，利用训练得到的权值制作了这个小程序，通过这个程序可以看出训练结果对数据集内的测试样本和训练样本都能进行很好的识别，但是对其他的手写字体识别就没有那么好了。(According to RBM depth network (Hinton, 2006) conducted MINST handwritten text classifier training, using the training to get the right value to produce this small program, this program can be seen through the results of the training data set and the training sample test samples are can be well identified, but on the other handwriting recognition is not so good.)

基于形状的模板匹配，用到了opencv，主体包含两部分，建模与金字塔搜索。(Based on the shape of the template matching, the use of opencv, the main body contains two parts, modeling and pyramid search.)

高光谱图像的地物分类技术，对光谱分析并对比检测出地景的算法(Fifth hyperspectral image classification techniques, algorithms for spectral analysis and contrast detect landscapes)

用粒子滤波实现视频中人头跟踪算法，有训练好的人头分类器。()

将传统的串行均值滤波算法修改为基于CUDA的并行的滤波算法(Modify traditional serial mean filtering algorithm based on CUDA parallel filtering algorithm)

完成图片的哈哈镜效果，如三角形，圆形，波浪形，弯曲等等(change the picture to any shape you want)

FIR滤波器的C语言实现 C语言源程序代码及解析(FIR filter, the C language realization C source code and analytical)

实现用帧差法画质心轨迹 高斯背景建模 可以运行的程序 opencv(Picture frame difference method implemented by the loci of Gaussian background modeling)

中国科学院长春光学精密机械与物理研究所 基于外极线约束条件的视频判读方法(Changchun Institute of Optics, Fine Mechanics and Physics epipolar constraint based on the video interpretation method)

图像增强的算法，但是有一定的局限性……但是对于一般比较典型的问题效果还是不错的……(Image enhancement algorithms, but there are some limitations ... but for a typical effect is good ...)