yanse

非常全面介绍了阴影检测算法以及如何去除阴影的算法，基本涵盖了该领域的所有知识点。(Is a comprehensive introduction to the shadow detection algorithm and how to remove the shadow of the algorithm, covering all the knowledge in the field point.)

影像仿射变换特征提取，结合 sift 等局部特征可以处理较大视角变化下的图像匹配。(This is an implementation of Hessian-Affine detector. )

There is a BCB program for digital image to establish ground truth image which test the original image.

基于马尔可夫随机场(MRF)、条件迭代算法ICM的图像分割源码(The MRF, the ICM-based image segmentation source code to source image segmentation based on the MRF ICM)

基于颜色特征向量的粒子滤波跟踪程序，修复了因粒子超过窗口范围出现ERROR的BUG，还包括了一些有助于粒子滤波学习的资料！！(color-based particle filter tracking, some bugs have been fixed, some information about PF Tracking is included too!!)

基于最小矩方判断灰度重心法计算光心位置的像素点，并输出结果。(Analyzing calculate pixel gray scale optical center position the center of gravity method, and the output torque based on the minimum square.)

说明：  双目测距bm算法，已经实现,基于vs2015+opencv3.2,测距精度很高，一米内精度一厘米误差(Binocular ranging bm algorithm, which has been implemented, is based on vs2015 + opencv3.2, which has high accuracy and one centimeter error in one meter)

同态滤波函数：使用频域处理方法的同态滤波函数，非高斯，巴特沃斯或指数型。对于较黑的图像效果更好(Homomorphic filtering function: the use of frequency domain methods of treatment homomorphic filtering function, non-Gaussian, Butterworth, or index. For more images and better)

对CANNY算子进行改进，其中详细的分析了CANNY算子的原理，已经应用改进的canny算子进行边缘检测，效果较好(CANNY operator improvements, including a detailed analysis of the principle of CANNY operator has been applied to improve the canny operator edge detection, better )

说明：  是一种纹理描述算子用于快速提取图像的纹理特征,应用于医学图像检索,场景分类等.(Is a texture description operator for rapid extraction of texture features, used in medical image retrieval, scene classification.)