imnoise2

立体视觉代码。主要过程包括：harris角点检测，zncc最优策略匹配，扩散得到更多特征对应，计算disparity和对应颜色值。另外提供了两幅图片以及对应的特征匹配的结果和disparity图像。(Three-dimensional visual codes. The main process includes: harris corner detection, zncc optimal strategy for matching, the proliferation of more features of the corresponding calculation of disparity and the corresponding color value. Also provides a picture and two corresponding feature matching results and disparity images.)

自行编写的逆投影变换，可实现图像坐标系到世界坐标系的转换，从实现鸟瞰变换。(the IPM ,which is completed by me, is able to achieve the transformation between the image coordinates and world coordinates )

说明：  双目测距bm算法，已经实现,基于vs2015+opencv3.2,测距精度很高，一米内精度一厘米误差(Binocular ranging bm algorithm, which has been implemented, is based on vs2015 + opencv3.2, which has high accuracy and one centimeter error in one meter)

说明：  MATLAB如何将图片切分成任意M*N块 在图像处理的时候，有时会不可避免的将图像分块来进行操作，但是图像的像素不可能每次都是可以被整除的，所以我们需要imresize函数对图像的尺寸进行修改，以便可以完成M*N块的分割(How can matlab cut the image into any m * n blocks? Sometimes it is inevitable to divide the image into blocks for operation. However, the pixels of the image can not be divided every time. So we need to modify the size of the image with the function of reduce to complete the m * n block segmentation)

说明：  图形界面计算psnr和ssim，直接输入原始图像和处理后的图像的psnr和ssim值，分别在Y通道和RGB通道(The graphical interface calculates PSNR and ssim, and directly inputs the PSNR and SSIM values of the original image and the processed image in Y channel and RGB channel, respectively)

一个外国人写的很好的meanshift聚类算法，有例程，可运行(Written by a foreigner meanshift good clustering algorithm, a routine can be run)

SAR图像理解与解译研究进展,电子学报,(SAR image understanding and interpretation of research)

rice大学CHOI写的一个关于小波域HMT模型的图像去噪算法。算法比较经典。对研究HMT模型的同行有很大参考价值。(rice university CHOI written on wavelet-domain HMT model for image denoising algorithm. Comparison of the classical algorithm. HMT model of research has great reference value counterparts.)

说明：  识别指纹端点分叉点以及奇异点，其中还包括指纹方向场和指纹梯度场的计算(It also includes the calculation of fingerprint direction field and fingerprint gradient field)

图像配准主要是通过两幅图像之间的互相关的最大值来确定配准的位置，从而在另一幅图像中显示配准的图像(Image registration is mainly through the maximum cross-correlation between the two images to determine the location of the registration and registered image, which is displayed in another image)