小波边缘检测

图像拼接小程序，使用直接对比的方法进行拼接并用加权平均的方法进行融合 (a simple program for image splicing )

第二代Curvelet变换及其在图像融合中的应用研究，第二代Curvelet较第一代有更好的发展前景(Second-generation Curvelet Transform and Its Application in Image Fusion of applied research, the second-generation Curvelet better than the first generation of the development prospects of)

相机标定工具箱，首先通过线性的方法求解出部分的摄像机参数，然后再利用非线性优化方法求出包括畸变系数在内的所有标定参数的最优解(Camera Calibration Toolbox for Matlab)

视觉slam十四讲，程序资料：g2o优化、cere拟合、特征点提取与匹配、点云拼接、opencv图像读取(Visual slam 14, program data: g2o optimization, cere fitting, feature point extraction and matching, point cloud mosaic, OpenCV image reading)

一本运用Visual C++ 的 数字图像处理的书里的源代码。非常好，初学者，很快可以上手.范例 3(Visual C++ for The digital image processing)

基于OPENCV的SIFT特征提取与匹配算法。包含完整的从图像高斯金字塔、DOG、空间极值点提取、关键点描述、KDtree匹配等关键步骤的全部函数实现，对全面深入理解Lowe的SIFT算法有莫大帮助。程序运行前须安装（1）OpenCV: http://opencvlibrary.sourceforge.net （2）SIFT: http://web.engr.oregonstate.edu/~hess/index.html，并配置其环境参数。(OPENCV the SIFT-based feature extraction and matching algorithm. Contains a complete Gaussian pyramid from the image, DOG, space extremum point extraction, description of key points, KDtree matching key step in the realization of the full function of the comprehensive and in-depth understanding of Lowe s SIFT algorithm of tremendous help. Program to run before the installation of (1) OpenCV: http://opencvlibrary.sourceforge.net (2) SIFT: http://web.engr.oregonstate.edu/ ~ hess/index.html, and configure the parameters of their environment.)

适用于高光谱图像目标检测的端元提取，该方法效果好(HYPERICAEEA Performs the ICA-Endmember Extraction Algorithm HyperIcaEea performs the ICA-EEA algorithm to generate abundance maps and find pure pixels. This function utilizes FastICA.)

说明：  区域生长法的图像分割，可以较好地提取脑部图像。(The region growing method can extract brain image well.)

生成归一化植被指数(NDVI)数据产品及图象产品的IDL源程序(that GIS is likely to have own itself advantages in sample selection, presentation of findings and future record collation (Department for Environment, Food and Rural Affairs, 2007). It indicates that to imply GIS software for mapping hedgerows may a reasonable and good choice.)

基于光线补偿的人脸检测。首先对图像进行光补，在此基础上结合肤色和发色进行人脸检测。(Light compensation based on face detection. First, the image light make up, in this based on the combination of color and hair color for face detection.)