AMCMF

运动检测 检测人体的头部位置、整体位置、滤除影子(Motion detection detect the body and head position, the overall position of the shadow filter)

本代码主要给出了激光粒度仪颗粒散射光强分布以及4种粒度反演算法，以及4种算法之间的比较。四种反演算法为：TSVD、Chaine、Tikhonov和l1正则化。(The code gives the Zetasizer particle scattering intensity distribution, and four kinds of particle inversion algorithm, as well as a comparison between the four algorithms. Four kinds of inversion algorithm is: TSVD, Chaine, Tikhonov and l1 regularization.)

基于样本的纹理合成，采取点替换的方法，可以合成brodatz类的纹理。采取的方法是先对图像分析，然后合成。(Texture synthesis based on samples taken at the replacement of the method of texture synthesis brodatz category. Approach taken is to firstly image analysis, and then synthesis.)

使用Matlab实现LMS自适应滤波，实现的主要功能有：读入原始音频，加入高斯噪声后用LMS自适应滤波滤除，比较原始音频，加入噪声音频和滤波后音频的区别。(Matlab achieved using LMS adaptive filtering to achieve the main features are: read the original audio, adding Gaussian noise is filtered using LMS adaptive filtering, comparing the original audio, add audio and noise filtered audio difference.)

卷积神经网络深度学习，可以识别多种图像，并自动解出所需要的结果(Deep learning of convolutional neural networks)

先进行图像预处理，比如Hough变换，边沿检测，还有差影法，模板提取 ，识别，跟踪，最后是模板匹配。(First image preprocessing, such as the Hough transform, edge detection, as well as poor shadow method, template extraction, identification, tracking, and finally the template matching.)

说明：  修改文件存储路径及相关参数,此代码可读取nc格式的文件(After modifying file storage path and related parameters,this code can read the data with the format of 'nc')

利用全卷积神经网络，实现图像的语义分割，基于tensorflow的keras可以直接运行(Using the full convolutional neural network to achieve semantic segmentation of images, keras based on tensorflow can be run directly)

1.支持多用户 2.支持自定义背景音乐，更个性化 3.可自定义浮动图片使留言本更漂亮(1. Support multi-user 2. Support for custom background music, more personalized 3. Customizable floating picture to make this message more beautiful)

说明：  阴影3D重建,对于人脸有效，基于yale2BDB(SFS, effective for face, based on yale 2 B database)