Form1

该VC++程序是基于CUDA架构的中值滤波程序，GPU加速，并且使用了opencv函数库(The VC++ program is based on the CUDA architecture median filtering program, GPU acceleration, and use the opencv library)

说明：  单目视觉中的位姿估算代码，基于正交迭代的OI算法，含有主程序和test代码，test代码可以看到位姿估算解的精度(Monocular vision pose estimation code, orthogonal iterative algorithm based OI, containing the main program and test the code, test code can see the accuracy of pose estimation Solutions)

可实现答题卡阅卷系统中的图像倾斜矫正，采用Hough进行直线检测，并根据所定位的特征直线位置计算倾斜角度，再进行图像旋转，得到校正结果(chieve answer sheet scoring system image tilt correction, performed using Hough line detection, and calculates the inclination angle in accordance with the target characteristic linear position, then rotate the image, corrected results)

canny边缘检测（自动设置Sigma的高斯平滑+Sobel(原始)+自适应直方图阈值）(canny edge detection (automatically set Sigma Gaussian smoothing+ Sobel (original)+ adaptive histogram threshold))

说明：  灰度图像的边缘检测，包含多种算子：sobel,prewitt,gaosi算子等，包括原理和各算子的区别(Edge detection of gray image includes many operators: Sobel, Prewitt, Gaosi, etc)

是一个很好的图像拼接程序，先用sift算法提取两幅图片的特征点再用算法筛去错配点，把匹配点用线连接起来，注意：主函数是match.m，运行主函数后在命令框输入运行match( image1.jpg , image2.jpg )(Is a good image stitching program, the first two pictures with the sift algorithm feature extraction algorithm and then weed out the wrong point with point, the match point with a line linking Note: The main function is match.m, after running the main function Enter the run command box match (' image1.jpg' , ' image2.jpg' ))

通过点云数据的格网化，建立格网分块索引，这是一种有效的点云数据存储方式。( Through the point cloud data grid , the establishment of a grid block index, which is a valid point cloud data storage. )

N-FINDR算法 端元提取 读取MNF降维后的影像数据 读取MNF降维后的影像数据( using N-FINDR end-member extraction)

用形态学方法和区域增长法进行MRI脑肿瘤图像分割(By morphology and region growing method for MRI brain tumor segmentation)

基于直接最小二乘法的椭圆拟合程序，比较实用，适用于图像处理，椭圆拟合方面研究的初学者(Direct least square method based on ellipse fitting procedure more practical for image processing, aspects of ellipse fitting for beginners)