sift-match

opencv安装指导文件，针对vs2010版本和opencv2.2版本(opencv installation guidance document for vs2010 version, and opencv2.2 version)

本程序采用迎风方案实现推广GAC模型的水平集方法，希望大家借鉴。(This procedure uses the wind to promote GAC model programs to achieve the level set, and I hope everybody' s reference.)

多基线干涉合成孔径雷达干涉条纹的获取，通过两幅单视复图像进行配准，获取干涉相位(Multi-baseline interferometric synthetic aperture radar interferometry fringes of the acquisition, through two single look complex image registration, access to intervention phase)

最小交叉熵图像分割源代码（原创）.相关参考文献见源代码。(source code for the minimum cross-entropy thresholding algorithm)

图像检测的算法，Faster R-CNN算法，先对整张图像进行卷积计算，然后通过感兴趣区域池化层(RoI Pooling Layer)将选择性搜索算法推荐出来的候选区域和卷积网络计算出的特征映射图进行融合，得到候选区域对应的特征矢量，这种共享卷积计算的操作极大地减少了卷积计算的次数。而且这些特征矢量的维度统一，方便后续的分类工作。通过感兴趣区域池化层处理卷积特征，并将得到的特征送往两个并行计算任务进行训练，分类和定位回归。通过这些方法和改进的框架，Fast R-CNN 用更短的训练和测试时长，取得了比 R-CNN 更好的效果(Faster R-CNN algorithm first convolutes the whole image, then fuses the candidate regions recommended by the selective search algorithm and the feature mapping maps calculated by the convolution network through the RoI Pooling Layer to get the corresponding feature vectors of the candidate regions, which greatly reduces the number of convolution calculations. Moreover, the dimension of these feature vectors is unified, which facilitates the subsequent classification work. The convolution feature is processed by the pooling layer of the region of interest, and the obtained feature is sent to two parallel computing tasks for training, classification and positioning regression. Through these methods and improved framework, Fast R-CNN uses shorter training and testing time and achieves better results than R-CNN.)

说明：  光流场经典算法-LK算法，通过两幅图像计算光流场。(Lucas – Kanade Motion Estimation using Pyramids)

利用Hough变换检测抛物线的算法。利用标准Hough变换从二值（黑白）图像中检测抛物线。(The use of parabolic Hough transform detection algorithms. Standard Hough transform from the binary (black and white) image in the detection of a parabola.)

在一个具有几个顶点的连通图G中，如果存在子图G 包含G中所有顶点和一部分边，且不形成回路，则称G 为图G的生成树，代价最小生成树则称为最小生成树。 (Has several vertices in a connected graph G, if there subgraph G ' contains all vertices of G and part of the side, and does not form a loop, called G' is a spanning tree of G, the minimum cost spanning tree is called minimum spanning tree.)

一种像素级指纹方向场估计算法以及对应的像素级奇异点检测算法。(A pixel-level orentation field estimation algorithm and the corresponding singular point detection method.)

有关三维模型的泊松算法的诸多应用的实现，包括泊松重建和泊松纹理混合算法，由原作者开发，非常可靠(C++ implementation for algorithms including poisson mesh and poisson texture stitching, it is shared by author with wide utilization)