extract-len-of-a-paper

图像匹配，将两幅不同的角度拍的图像的对应点都找到(Corner detection)

盲去卷积复原图像的示例翻译，加入了中文标注，效果较好(deconvblind)

这是我的大作业，作用是测试车流量并计数，使用高斯模型对车辆进行定位分割，已验证成功内附测试视频(This is my big job, the role is to test the traffic flow and counting, the use of the Gauss model of the vehicle positioning and segmentation, has been verified successfully included test video)

运用Perona-Malik各向异性扩散模型对3d图像进行平滑去噪。此方法可以有效的抑制高斯噪声，并保护图像的边缘特征。(The use of Perona-Malik anisotropic diffusion model for smooth 3d image denoising. This method can effectively suppress Gaussian noise and to protect the image of the edge feature.)

retinex算法的一个源代码，用于图像增强处理，能比较好的提高对比度(retinex an algorithm source code for image enhancement, to better enhance the contrast)

说明：  针对散焦图像编写的复原程序，复原效果不错，会产生振铃效应！(For the preparation of defocused image restoration program recover good results, will produce ringing effects!)

里面包含kalman，扩展kalman，无迹kalman，粒子滤波，无迹粒子滤波等源码的实现。(upf--The Unscented Particle Filter )

基于稀疏分解的欠定盲源分离，要求信号具有一定的稀疏性。(Based on sparse decomposition due to blind source separation, the signal has a certain sparseness.)

使用选择式掩膜平滑方法进行数字图像的边缘提取(Select the type mask smoothing method using digital image edge extraction)

SIFT，即尺度不变特征变换（Scale-invariant feature transform，SIFT），是用于图像处理领域的一种描述。这种描述具有尺度不变性，可在图像中检测出关键点，是一种局部特征描述子。 该方法于 1999 年由 David Lowe 首先发表于计算机视觉国际会议（International Conference on Computer Vision，ICCV），2004 年再次经 David Lowe 整理完善后发表于 International journal of computer vision（IJCV） 。截止 2014 年 8 月，该论文单篇被引次数达 25000 余次。(来自百度)(SIFT, namely Scale-invariant feature transform (SIFT), is a description used in the field of image processing. This description has scale invariance and can detect key points in images. It is a local feature descriptor. This method was first published by David Lowe at the International Conference on Computer Vision (ICCV) in 1999 and published in the International Journal of Computer Vision (IJCV) in 2004 after it was sorted out and perfected by David Lowe again. As of August 2014, more than 25,000 citations have been cited. (From Baidu))