基于复杂背景的人体手势识别，以及对它的鲁棒性的测试

数字图像处理中基于vc++的颜色空间变换源程序-digital image processing based on vc color space conversion source

这是一个关于图象融合的程序，利用该程序可以实现任意图象的匹配与融和。-This is a picture of the integration process, the use of the program could realize arbitrary image matching and integration.

基于Graph Cuts计算机视觉立体视图匹配算法-Computer Vision Graph Cuts Based on three-dimensional view matching algorithm

语音编码文档，将ILBC代码专为硬件时需进行操作及对代码进行优化-Speech coding documentation, will be ILBC code specifically for the hardware required to operate and optimize the code

基于环形颜色直方图的图像检索算发，比传统方法好用的多-Annular color histogram-based image retrieval count hair, easy to use than the more traditional methods

此源代码为图像处理提供各种基本操作，可以为做实验节约很多时间。-This source code for image processing to provide a variety of basic operations, we can save a lot of time doing experiments.

基于小波变换的图象去噪 Normalshrink算法-based on wavelet transform algorithm for image denoising Normalshrink

影像的读入，简单的几何变换，影响增强，边缘检测，等，用于学习影像的操作。-Reading images into a simple geometric transformation, the impact of enhanced edge detection, etc., for learning the operation of the image.

Dual Image Registration Software kit

基于轮廓的形状特征提取 1)该程序在VC＋＋6.0下编译通过； 2)测试图片说明：（图像路径名请在main函数中fp=fopen("1.bmp","rb")中设定） 64×64 24bit BMP： 图像1、2用来验证旋转图像识别；图像1、3用来验证平移图像识别；图像1、4用来验证缩放图像识别； 120×120 24bit BMP： 图像5、6用来验证平移图像识别；图像7、8用来验证旋转图像识别； 3)结果说明： MCS.dat保存了Mountain Climbing Sequence。 -Based on the outline of the shape feature extraction 1) the program in VC 6.0 under the compiler through 2) Test Photo Caption: (image path name in the main function fp = fopen (1.bmp, rb) set) 64 × 64 24bit BMP: image 1,2 to verify rotating image recognition image 1,3 to verify translational image recognition image 1,4 to verify scaling image recognition 120 × 120 24bit BMP: images used to verify translational 5,6 image recognition images 7,8 rotate images used to verify identification 3) The results indicate that: MCS.dat preserved Mountain Climbing Sequence.