celledge7

可对lytro相机所得到的光场图像进行各种处理(The light field image obtained by lytro camera can be processed in various ways.)

自己编写的广义霍夫变换，里面有测试图片，可以作为学习霍夫变换的入门(I have written a generalized Hough transform, there are test images, the Hough transform can be started as a learning)

计算机图形学图像处理学习内容，附带PPT教程里面的代码。(the ppt for Digital Image)

a lane dtection method using hough-transform

一种基于匈牙利算法和相关向量机的点模式匹配算法，可以有效解决点集的刚性和非刚性变换的参数求解问题。(A point set registration method based on Hungarian Algorithm and RVM which can well solve the parameters of rigid and non-rigid transform.)

图像去雾 暗原色 导向滤波 最好的去雾算法 何恺明的matlab代码的opencv的实现(haze removal guided image filtering the best algorithm to fog )

图像分类程序，此图像分类采用 SIFT + Kmeans 聚类的方法，然后调用 MLP 对其特征进行分类处理，速度实现比较快，正确率高(Image classification procedures, the use of this image classification method SIFT+ Kmeans clustering, and then call the MLP classification of its features, faster speed to achieve the correct rate)

DIC测试，适用于二维数字图像相关法，采用的ZNSSD，输入图像需为灰度图像，输出仿射矩阵(Measuring strain in samples which are too small, big, compliant, soft or hot are typical scenarios where non-contact techniques are needed. A technique which can cover all that and also can deal with complicated strain fields in structures or structural materials is the Digital Image Correlation. With this technique, strain can be calculated from a series of consecutive images with sub pixel resolution as will be shown in the following chapters.)

一种基于双立方插值的新型图像插值算法，较好的保存边缘信息(Based on bicubic new image interpolation algorithm, the preservation of edge information better)

基于contourlet的图像融合，低频分量在融合过程中使用加权平均的融合规则(Contourlet-based image fusion, the low frequency components in the integration process using the weighted average of the fusion rules)