mmn

使用cnn提取图像特征，然后用SVM分类，此处没有给出训练集，另外imagenet-caffe-alex部分代码需要注意，需要下载的话把注释掉的代码打开(Using CNN image feature extraction, and then use the SVM classification, there were no given training set, also need to pay attention to imagenet-caffe-alex part of the code, you need to download the commented code open)

MCA，形态成分分析，广义图像处理，图像分解，图像降噪处理。(gmcalab fgmca.m imnb.m sparse_noisy_examples.m)

推荐火灾检测的参考文献，有火焰多个特征提取的算法。(the article is about how to detect fire，using many algorithms to get the features of fire.)

今年刚更新的最新的工具箱SaliencyToolbox2.3.zip，做图像处理视觉注意力机制相关研究必备的工具。(This year has just updated the latest toolbox SaliencyToolbox2.3.zip, do image processing of visual attention mechanisms related research is an essential tool.)

基于matlab的时频率分析工具箱，包括了小波变换，WVD，SPWVD等等好多的时频分析方法(Matlab-based time-frequency analysis toolbox, including a wavelet transform, WVD, SPWVD so many time-frequency analysis method)

该源码为目标检测和跟踪算法 包括： 1. 静态背景下的背景预测法目标检测 2. 静态背景下帧间差分法目标检测 3. Mean Shift目标跟踪方法 4. 重心多目标跟踪方法 (The source code for target detection and tracking algorithm include: 1 Static background background prediction target detection 2 Static background frame difference method for target detection 3. Mean Shift tracking method 4 multi-target tracking methods focus)

本程序主要实现暗背景下含高斯噪声的光斑滤除噪声，实现光斑中心提取。采用了sobel算子和梯度倒数加权滤噪、最大类间方差法（Otsu）确定阈值、质心法提取光斑中心坐标。(The program is mainly realized under a dark background Gaussian noise spot containing noise filtering, to achieve spot center extraction. Using sobel operator and gradient inverse weighting filter noise, Otsu method (Otsu) to determine the threshold, the center coordinates of the spot centroid extraction.)

基于Gabor的特征提取,gabor滤波器的构造 包含中科院的虹膜库(base on gabor zhongkeyuandehongmoku)

OMP算法的matlab的详细描述，对学习正交匹配追踪算法和稀疏分解是很有用的(the OMP algorithm is describled by matlab,and it is very useful for us to study OMP algorithm and sparse decomposition.)

基于物体灰度图像的物体三维形状恢复，修改图形路径和光源方向即可(Based on the three-dimensional shape of the object to restore the object grayscale image, modify the graphics path and direction of the light source can be)