数学建模运动模糊图像的复原

Jianchao Yang 的基于稀疏表示的单幅图像重建的原始代码，先将高低训练图像分块，再将块训练成高低字典，将测试图像映射到低字典上，得到系数，再乘以高子典就得到最后的图像。对学习超分辨率同学的参考作用很大。(This is the original matlab code for super resolution by Jianchao Yang 。The method is sparse represent based on the overcomplete dictionary。)

说明：  在产生混沌伪随机序列的混沌映射中，Logistic映射(Logistic Map)是一种简单而经典的映射。其中u称为分形参数，随着u逐渐增大，出现倍周期分岔现象。可用于密文域图像加密。(Logistic Map is a simple and classical mapping for generating chaotic pseudo-random sequences. U is called fractal parameter, and with the increase of u, period doubling bifurcation occurs. It can be used for image encryption in ciphertext domain.)

本程序通过水平和垂直灰度投影，框选出灰度面部图像中的眼睛区域。(This program is used to locate eye region in gray facial image by means of horizontal and vertical projection.)

程序为读取图片并对图片进行边缘检测，先采用高斯滤波进行去噪，然后通过求梯度运算检测，非极大值抑制边缘检测，双阈值检测这三种方法，分别对图片进行边缘测，并对边缘检测后显示的图片进行对比。此程序对边缘检测有很大的参考价值。(Program to read the image and image edge detection, Gaussian filter to remove noises, and then detected by calculating the gradient operation, the non-maxima suppression edge detection, dual-threshold detection of these three methods, namely the picture edge test, and the rear edge detection to display image comparison. This program is great for edge detection reference value.)

说明：  利用稀疏矩阵思想进行图像的超分辨率重建，效果较好。(Thinking of using sparse matrix for image super-resolution reconstruction, the effect is better.)

由于前台开发人员经常需要通过RGB值指定各种各样的颜色，而通过这几个图片就可以马上找到某个颜色的RGB值，从而更快地编写网页(Foreground developers often need a variety of colors through RGB values ​ ​ specified by these pictures can immediately find the RGB value of a color, so to write Web pages faster)

主成分分析算法（PCA），可用于降维，也可用于处理图像相关性问题，提取主成分，分析图像细节信息和主要成分，用于图像压缩也可以(Principal component analysis algorithm (PCA), can be used for dimensionality reduction, can also be used to process images related issues, extracted principal component analysis and main component of image detail information, to be used for image compression)

对最大类间方差法进行改良，引入最小类内方差，进一步提高分割效果(For improved Otsu method, introducing the minimum interclass variance, further improve the segmentation effect)

人耳识别，包含图像预处理，图像特征点的提取，比较实用(Ear recognition)

二维码扶正,对不同光照图片都有较好结果，涉及几种不同的二值化方法，采用霍夫检测直线和双线性变换扶正二维码，内含的7张图片都能扶正(Two-dimensional code righting, have a better picture of the different light result, involving several different binarization method using Hough transform to detect linear and bilinear righting two-dimensional code, embedded image 7 can be righting)