circledetect

生成avi视频截图序列。设置截图时长，截图频率，截图类型的参数。(Generate avi video sampling sequence.Screenshot duration, frequency and types of parameters are allowed to set freedom.)

采用了Harrtraining和离子滤波算法实现了拥挤人群人头计数，非常有用。VC++和Opencv开发，采用Oepncv中的Harrtraining训练一个分类器即可运行(Human detection and tracking in high density crowds is an unsolved problem. Standard preprocessing techniques such as background modeling fail when most of the scene is in motion. We integrate human detection and tracking into a single framework, and in- troduce a conrmation-by-classication method to estimate condence in a tracked trajectory, track humans through occlusions, and eliminate false positive detections.)

说明：  实现平滑0范数压缩感知算法，主要对二维图像进行压缩感知重构。自行设置图像的采样数目、添加图像即可运行，无需修改。(Realizes a smooth 0-norm compressed sensing algorithm, mainly for compressed sensing reconstruction of two-dimensional images. Set the sample number of the image by yourself, add the image to run, no need to modify.)

用于图像拷贝检测的图像匹配，提取图像的SIFT特征点，构成图像的特征。(For image copy detection of image matching, image SIFT feature point extraction, and then sift feature point the image characteristics)

图像修复是图像复原研究中的一个重要内容, 它的目的是根据图像现有的信息来自动恢复丢失的信息，其可以用于旧照片中丢失信息的恢复、视频文字去除以及视频错误隐藏等。(Image restoration image restoration is an important part of the study, its purpose is to present the information according to the image to automatically recover lost information, which can be used for missing information in the old photo restoration, video, text and video error concealment removal .)

利用蝶形算法类似于FFT，利用先行再列实现二维DCT。速度很快。包含反DCT的代码。(similar to the use of butterfly FFT algorithm, the first re-use of 2D DCT out. Very fast. Contain anti- DCT code.)

图像处理技术中用于图像聚类的一段程序，即将区域中具有相同灰度或其他属性的区域提取出来。 (image processing technology for image clustering section of procedures, upcoming region with the same intensity or other attribute extraction from the region.)

说明：  图像处理，通过matlab程序实现视频插帧的功能(Image processing and video frame insertion through MATLAB program)

对红外图像进行增强，采用多小波分解进行分层增强(Infrared image enhancement, wavelet decomposition of multi-layered enhancement)

2015年最新国外算法，基于加权视觉词典的人群密度估计(crowd entimatin by a weighted bag of visual word model)