KNN

基于Hough 变换的SAR 图像舰船尾迹检测方法。该文引进了K均值聚类方法,将这些点集归类,归类后的每一类的类心能很好的代表一条尾迹,实验结果表明方法是有效的。(Hough transform based on SAR images of ship wake detection. The introduction of the K-means clustering method to classify these points, is classified in each category after the class on behalf of the heart can be a good wake, experimental results show that the method is effective.)

说明：  简单的Halcon 找圆算法，首先进行圆检测(识别圆、圆拟合)，再找出坐标位置(A simple Halcon algorithm for finding circles firstly conducts circle detection (identifying circle and circle fitting), and then finds the coordinate position)

使用二維的傅立業轉換，在以反傅立業的方式，將結果以影像話方式呈現。(Industry use of two-dimensional Fourier transformation, anti-Fu Li industry in the way of results, then the way to show images.)

在对话框中实现splitter(achieve splitter)

非压缩视频图像帧内水印的密写与复密写研究，源码和论文材料(Video Watermarking Steganography)

matlab的四种边缘提取办法比较（代码）(the four edge detection matlab comparison approach (code))

PCL是一个类似于OpenCV的开源库，提供了很多三维点云的处理功能，其中就包括点云拼接。在三维扫描项目中，需要利用点云拼接方法将多次扫描得到的点云数据拼合成一个整体。在本例中，用一幅包含人脸的深度图像去匹配之前获得的人脸模板，这将使我们能够确定人脸在场景中的位置和方向。(PCL is an open source library similar to OpenCV, which provides a lot of processing functions for three-dimensional point clouds, including point cloud mosaic. In the three-dimensional scanning project, it is necessary to use point cloud mosaic method to combine the point cloud data obtained from multiple scans into a whole. In this example, a depth image containing face is used to match the face template obtained before, which will enable us to determine the position and direction of the face in the scene.)

三维数据去噪处理，一维。二维数据去噪也可以使用(3D data denoised)

使用帧差法、均值法和混合高斯模型对背景进行去除、提取前景信息。来源于2017年研究生数学建模比赛D题。(The background information is removed and the foreground information is extracted by frame difference method, mean value method and hybrid Gauss model. D graduate from the 2017 mathematical modeling contest.)

主要是一些经典的视觉跟踪算法的MATLAB代码，tracker包括CREST,CSK,CT,ECO,KCF,LOT,SCM和Struck。(Visual object tracking,include CREST,CSK,CT,ECO,KCF,LOT,SCM and Struck.)