mean-shift

sift特征提取，提取dense-sift特征，快速，易懂，很不错的(sift feature extraction, extraction dense-SIFT features, fast, easy to understand, very good)

说明：  使用python中的opensilde库对tiff格式图片进行分割。(TIFF format image segmentation.)

程序实现了三种去噪算法：BLS-GSM/BM3D/NLM，评价指标为PSNR/SSIM.(Program implements three de-noising algorithm: BLS-GSM/BM3D/NLM, with evaluation of PSNR/SSIM.)

说明：  利用python写的简单降噪程序（根据一个点A的RGB值，与周围的8个点的RBG值比较，设定一个值N（0

通过纹理方向估计消失点，通过消失点来检测道路区域。 该程序应用到了基于纹理的张量投票算法来估计消失点；通过纹理方向和颜色信息，生成用于估计两个最主要道路边界的直方图；最后将道路区域定义为两个检测到的道路边界之间且低于估计的消失点的区域(The vanishing point is estimated by the texture direction, and the road area is detected by the vanishing point. The program applies a texture-based tensor voting algorithm to estimate the vanishing point; through the texture direction and color information, generates a histogram for estimating the two most important road boundaries; finally defines the road area as two detected road boundaries. An area between and below the estimated vanishing point)

本程序将SIFT与边缘结合起来用来做图像配准 先用SIFT检测关键点 用CANNY算子进行边缘检测 进而优化选取最佳匹配点 进行向量匹配 图像配准。 (The program will combine the SIFT and the edge of image registration is used to make the first detection of the key points with SIFT operator with CANNY edge detection and then select the best match point optimized vector matching image registration.)

几种经典的图像平滑去噪算法程序，其中包括小波软阈值和硬阈值处理。还附送三个图像质量评价程序(several classic image denoising smoothing algorithm procedures, which include thresholding wavelet soft and hard threshold processing. Also attached three image quality evaluation procedures)

图像修复，使用曲率方法，热传递。边缘修复非常好，参考某博士论文写。(Image restoration, using the curvature method, heat transfer. Edge of the repair is very good, the reference of a doctoral dissertation.)

说明：  有四个M文件，可以用于图像处理中的图像细化，有较好的效果(There are four M documents, can be used for image processing of image thinning, there are good results)

基于高阶互信息的医学图像配准算法 基于高阶互信息的医学图像配准算法代码(High-level mutual information based medical image registration algorithm based on mutual information of high-end medical image registration algorithm code)