FG

说明：  行为识别3D卷积神经网络,ucf或者KTH数据集都可以使用(3DCNN for action recognition)

对于原始图像 < f5.tif > ，实现下列结果。 a.用平均滤波器平滑原始图像得到的模糊图像 b.用拉普拉斯算子增强上述a得到的模糊图像 c.用拉普拉斯算子增强原始图像 (For the original image <f5.tif>, to achieve the following results. a. smoothing filter with an average original image was blurred b. using Laplacian enhancement get a blurred image above c. using Laplacian enhancement of the original image)

在张正友的基于平面方格的摄象机标定中,单应矩阵是关键的一步,本程序是效率精度比较高的源码(Zhang is the Friends of the plane on the box camera calibration alone should matrix is a crucial step, The procedure is relatively high efficiency accuracy of source)

一个师兄编写的图像配准程序，其中sift代码匹配以及错误点检测值得参考学习。(A brother write image registration procedures, which sift code matches the reference point of error detection is worth learning.)

说明：  opencv基础教程，对于初学者来说是个不错的选择。(opencv basic tutorial for beginners is a good choice.)

有时为了方便处理，我们需要对大量的图像格式进行转换，本程序就是对图像格式批量转换。(Sometimes in order to facilitate the image processing, we need to tranfer the format of a lot of images.This program is to tranfer the image format in batches.)

此算法是图像之间的转换，二值图，RGB等的相互转换源代码(This algorithm is the conversion between the images, the two value of the map, RGB, and so on the conversion of the source code)

一篇关于总结了行为识别各种资源的论文，涵盖了通常的数据集、兴趣点检测，特征提取的方法以及识别检测的算法(A paper that summarizes the various resources of behavior recognition, covering the normal data set, the interest point detection, feature extraction methods and the identification detection algorithm)

压缩感知中的梯度投影重构算法，经修改可以运用与单像素相机成像(Gradient projection reconstruction algorithm of compressed sensing, modifications can be used with single pixel camera imaging)

说明：  Jpeg算法源代码，包含jpeg的各种算法实现的vc++的源代码(Jpeg algorithm source code, including the various algorithms jpeg of vc++ source code)