Image Classification Based on Neural Network

GLCM feature 1 for image

matlab gui图像处理，很全很实用，值得大家慢慢学习(matlab gui image processing, it is all very practical, worth learning slowly)

全变分去噪 J=tv(I,iter,dt,ep,lam,I0) 输入: I - 灰度图像, iter - 迭代次数 [1], dt - 时间步长 [0.2], ep - 提升参数epsilon [1], lam - 保真项 lambda [0], I0 - 输入噪声图像 [I0=I] ([]中的是缺省值)( 全变分去噪 J=tv(I,iter,dt,ep,lam,I0) 输入: I - 灰度图像, iter- 迭代次数 [1], dt - 时间步长 [0.2], ep - 提升参数epsilon [1], lam - 保真项 lambda [0], I0 - 输入噪声图像 [I0=I] ([]中的是缺省值))

VC++图像平滑处理+程序源码，实现图象的平滑（去噪声），锐化功能，测试时请在C盘下放一个测试用的Bmp图片，路径：C: est.bmp，没有的话没法测试。如上图所示是水平平滑处理，可以对比下，不过感觉颜色失真了，有兴趣的看一下。命令行编译过程如下： 　　vcvars32 　　rc bmp.rc 　　cl smooth.c bmp.res user32.lib gdi32.lib(VC++ image smoothing+ program source code, to achieve a smooth image (to noise), sharpening function in the C drive when testing a test of decentralization Bmp picture, path: C: test.bmp, no words can not test . Is shown above the level of smoothing, you can compare the next, but the feeling of color distortion, interested look. Command-line compiler as follows: vcvars32 rc bmp.rc cl smooth.c bmp.res user32.lib gdi32.lib)

改变图像灰度，增强图像中需要的信息，可利用与分割前处理(Change the image gray level, increase the information)

目标分类（流形学习slle+支持向量机svm）(Target classification (manifold learning slle+ support vector machine svm))

hessian矩阵在图像处理中的应用。根据图像所得函数，建立相应的hessian矩阵(hessian matrix in image processing applications. The resulting function according to the image, to establish the corresponding hessian matrix)

根据论文Curvature Scale Space Corner Detector with Adaptive Threshold and Dynamic Region of Support算法写的角点检测(err)

说明：  基于遗传和蚁群混合算法的图像配准算法 需要利用matlab遗传工具箱(Based on genetic and ant colony algorithm for hybrid image registration algorithm need to use genetic toolbox matlab)

高斯模糊是一种图像滤波器，它使用正态分布(高斯函数)计算模糊模板，并使用该模板与原图像做卷积运算，达到模糊图像的目的(Gaussian blur filter is a kind of image, it uses the normal distribution (gaussian function) to calculate fuzzy template, and use the template and the original image for convolution operation, achieve the goal of blurred image )