DLA

说明：  用于立体匹配的三种算法，显示视差图，matlab代码(Three algorithms for stereo matching, display parallax map, matlab code)

基于MATLAB的SOM神经网络设计源程序(MATLAB-based SOM neural network design source)

人工选择聚类中心，用k-means聚类方法对图像进行分割，效果不错的(artificial selection cluster center, k-means clustering method for image segmentation, the good results)

实现图像的分块压缩感知重建，DCT变换，OMP重构算法(Image block compressed sensing reconstruction, DCT transform OMP reconstruction algorithm)

thin plate spline（薄板样条）模型，将已知图像通过非刚性变换得到所需要的图像，在图像配准中非常有用。(thin plate spline (thin plate spline) model, will be known through non-rigid image transformation to be necessary by the images of image registration in very useful.)

MICC-F220.zip matlab ..............

压缩文件是CNN卷积神经网络图像处理的数据库mnist_uint8.mat(Compressed file is the database of CNN convolution neural network image processing mnist_uint8.mat)

对叶片进行预处理和提取部分特征：1.均值去噪 2.k均值聚类提取复杂背景下的叶子图片 3.填充孔洞 4.去叶柄 5.提取轮廓 6计算纵横轴比、面积凹凸比、周长凹凸比、球形性、圆形性、偏心率、形状参数和矩形度等特征值并进行画图。( Pretreatment of leaves and extract some of the characteristics: a mean denoising 2.k means clustering leaf extract complex background picture fill holes 3 4 5 contour extraction petiole 6 to calculate the vertical and horizontal axis ratio, convex area ratio.. convex perimeter than spherical, circularity, eccentricity, and rectangular shape parameter values ​ ​ and other characteristics of the drawing.)

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)

说明：  基于c#运动控制及图像处理项目源码，有兴趣可以看看(Based on C# motion control and image processing project source code)