数学建模运动模糊图像的复原

用粒子滤波实现视频中人头跟踪算法，有训练好的人头分类器。()

用matlab仿真，研究不同阈值降噪后的效果，使用penalty阙值降噪后信号，使用Birge_Massart阙值降噪后信号，使用缺省阙值降噪后信号。(Using matlab simulation to study the different threshold noise reduction effect, the use of penalty threshold denoised signal after noise reduction using Birge_Massart threshold signal, use the default threshold signal after noise reduction.)

说明：  fluent udf dpm particle diamter

图象分割程序：先用梯度函数进行边缘检测,然后二值化后分割图象(Image segmentation process: first the gradient edge detection function, and then after binarization image segmentation)

人脸检测 通过肤色建立模型，和计算欧拉数开运算等一系列处理可以检测出同一张图片中的多个人脸！效果不错哦(face detection through color model building, and calculation of Euler number opened a series of processing operations can be detected with a picture of the personal face! Oh, good results)

说明：  图像融合指标，综合熵、空间频率、互信息等等，一共7种。在图像处理后的质量评价方面很有用(Fusion index, comprehensive entropy, spatial frequency, mutual information, etc., a total of 7 species. In the image processing is useful for quality assessment)

实现文章“Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries”中的降噪算法(Realize the article " Image Denoising Via Sparse and Redundant Representations Over Learned Dictionaries" in the noise reduction algorithm)

HOG adaboost pedestrian-detect

YUV视频文件二维运动估计中的三步法源程序，可以直接运行！(Three Step Search)

矩阵卷积的几种版本：conv函数，矩阵乘法形式，纯矩阵乘法形式(Several versions of matrix convolution: convolutional vision, matrix vision, only-matrix vision)