winImageCompletion-master

一段很好的图像放大缩小程序，解码后可以运行。(A very good image magnification narrow program, the decoding can run. )

图像处理源程序，实现二值化变换，经一幅图像灰度化后，变成二值化图像，包括大津法、最大类间方差法、一般阈值法、及文本识别最有用的Niblack法(Image processing program and realized binarization transformation, after by a gray-scale image into binary image, including the Otsu method, Otsu method, the general threshold, and text to identify the most useful Niblack method)

一种基于背景减法和帧差的运动目标检测算法(Based on background subtraction and frame difference moving target detection algorithm)

本程序为图像锐化的经典程序，源码里有详细的标注。(the procedures for sharpening images of the classic procedures are detailed source of tagging.)

包含滤波模板，形态学运算，边缘检测结合经自己修改过的hough变换编写的vc程序，成功检测机场跑道和任何图像中最明显的现状特征，经测试，检测精度高。对学习vc编程用于图像处理有帮助。(hough transform based image processing package which is useful for detecting airfield like straight line features.)

（1）在工具栏中点击“打开”功能出现文件打开窗口，用户选择一幅bmp图片后在指定区域居中显示该图片，若图片尺寸小于指定区域则按原始大小显示；若大于则按合适比例显示。 （2）鼠标指针在图片上停留时显示为十字形，同时在控件下方固定位置显示鼠标所在的图像x坐标和y坐标（如：50，100），以图片左上为坐标原点。 （3）在工具栏中可以点击控件实现在图片上绘制一次直线、矩形或圆等图形，在绘图过程中要求图片和图形一直可见，不能出现刷新痕迹，绘制的图形可以任意修改形状。 （4）可以任意选择图形线条的类型（虚线或直线）、颜色（三种）及粗细（三种），要求选择方式可以叠加，绘制过程要求可任意连续执行。 （5）在工具栏中点击“保存”功能出现文件保存窗口，用户可以选择保存路径来保存当前图片操作结果（包括图片和图形）。 （6）点击“关闭”功能，关闭当前图片，若操作结果尚未保存则必须弹出对话框提示用户进行保存。 ((1) in the tool bar, click on Open function has the file open window, the user selects a bmp image to center the picture in the designated area, if the image size is smaller than the specified area press the original size if more than press the appropriate scale display. (2) is shown as a cross when the mouse pointer over the image, while displaying an image x and y coordinates of the mouse cursor in a fixed position below the control (eg: 50,100) to image the upper left of the coordinate origin. (3) in the toolbar you can click controls to achieve a draw straight lines, rectangles or circles and other graphics on the picture, the requirements of pictures and graphics in the drawing process has been seen, it can not appear refreshed traces drawn graphics can be modified shape. (4) can choose the type of graph line (dashed or straight), color (three kinds) and thickness (three kinds), asked to choose the way can be superimposed, can arbitrarily draw process requires continuous exe)

说明：  使用gpu、cpu并行进行sift算子计算匹配，能够在原来的基础上加速处理，但对显卡要求较高，具体环境配置使用方法可以参照mannual(SiftGPU is an implementation of SIFT [1] for GPU. SiftGPU processes pixels parallely to build Gaussian pyramids and detect DoG Keypoints. Based on GPU list generation[3], SiftGPU then uses a GPU/CPU mixed method to efficiently build compact keypoint lists. Finally keypoints are processed parallely to get their orientations and descriptors. SiftGPU is inspired by Andrea Vedaldi s sift++[2] and Sudipta N Sinha et al s GPU-SIFT[4] . Many parameters of sift++ ( for example, number of octaves, number of DOG levels, edge threshold, etc) are also available in SiftGPU. The shader programs are dynamically generated according to the parameters that user specified. SiftGPU also includes a GPU exhaustive/guided sift matcher SiftMatchGPU. It basically multiplies the descriptor matrix on GPU and find closest feature matches on GPU. Both GLSL and CUDA implementations are provided. )

源码包括了常用的图像处理方法，代码详细，适合初学者(Source code, including the commonly used image processing method, the code in detail, suitable for beginners)

说明：  SAR影像单视数据转为多视数据，数据间相互转化，快捷处理SAR数据(Single-look SAR image data into a more visual data, data inter-conversion, fast processing SAR data)

图像处理的IPP库使用手册，包括每个函数的使用说明，以及参数的意义。(IPP library of image processing using the manual, including instructions for each function, as well as the significance of the parameters.)