石鑫华视觉网-中科院自动化所 计算机视觉课件

基于稀疏表示的多光谱图像融合源代码，用于pansharpening(sparse-based mutispectral image fusion)

使用经典边缘检测算子进行边缘检测，加入噪声的图像边缘检测结果和未加噪声的图像检测结果进行对比分析，默认阈值设置的图像检测结果和手动设置阈值的图像检测结果进行对比(Edge detection using classical edge detection operator)

图形voronoi图的自动生成算法，用vc做，是基于一般图形voronoi图的算法，结合数字图形处理里的轮廓跟踪算法，使一般图形voronoi图能够根据不同的生成元总动生成。(Graph voronoi diagram automatic generation algorithm, using vc do is based on the general graph voronoi diagram algorithm, combined with digital image processing in the contour tracking algorithm, the general graph voronoi diagram to the total depending on the generator automatically generated.)

关于超分辨率图像恢复的凸集投影算法，运行结果正确。(code about super-resolution based on POCS)

归一化互相关算法是传统的图像匹配算法,针对传统图像匹配算法运算量大、速度慢的缺点,提出了一种基于小波变换的多尺度图像匹配算法。(Normalized cross-correlation algorithm is the traditional image matching algorithm, proposed a wavelet transform-based multi-scale image matching algorithm for the large amount of computation of traditional image matching algorithm, slow disadvantage.)

说明：  根据：一种可识别破碎图形的特殊广义Hough变换方法 论文，在matlab上做的实验源码，有模版生成，广义哈夫变换代码(In accordance with: A broken identifiable graphics special generalized Hough transform methods papers, in matlab to do experiments on the source, there are template generation, generalized Hough transform code)

ViBe算法是由Olivier Barnich 和 Marc Van Droogenbroeck在2011年提出的一种背景建模方法。该算法采用邻域像素来创建背景模型，通过比对背景模型和当前输入像素值来检测前景.(ViBe algorithm is a background modeling method proposed by Olivier Barnich and Marc Van Droogenbroeck in 2011. The neighborhood pixels are used to create the background model, and the foreground is detected by comparing the background model and the current input pixel values)

本程序包是基于多分辨率分析的图像融合算法，其中包括基于区域的融合算法。对应程序有相关的文章，是学习图像融合最有用的资料。(This package is based on the analysis of multi-resolution image fusion algorithms, including region-based fusion algorithm. Procedures related to the corresponding articles, image fusion is to study the most useful information.)

说明：  给定两张图片，利用ssim来计算两张图片他们之间的ssim相似度(compute ssim between two images)

寻找最低像素值的matlab实现：通过图像矩阵的循环查找，能实现最低像素值的快速寻找(Matlab find the lowest pixel value to achieve: to find the cycle through the image matrix, to achieve the minimum pixel value Quick Find)