pre_processing

压缩感知，稀疏表示采用小波基表示，压缩测量采用随机高斯矩阵，重构算法是omp重构(Compressed sensing, sparse representation using wavelet representation, compression measurements using random Gaussian matrix remodeling reconstruction algorithm is omp)

说明：  利用matlab对已知多对控制点的情况下进行两张图像校正，该算法能够取得较好的结果(Using matlab to correct two images when multiple pairs of control points are known.The algorithm can get good results.)

混合高斯(Mixture of Gaussian, MOG) 背景建模算法和Codebook 背景建模算法被广泛应用于监控视频的运动目标检测问题, 但混合高斯的球体模型通常假设RGB 三个分量是独立的, Codebook 的圆柱体模型假设背景像素值在圆柱体内均匀分布且背景亮度值变化方向指向坐标原点, 这些假设使得模型对背景的描述能力下降. 本文提出了一种椭球体背景模型, 该模型克服了混合高斯球体模型和Codebook 圆柱体模型假设的局限性, 同时利用主成分分析(Principal components analysis, PCA) 方法来刻画椭球体背景模型, 提出了一种基于主成分分析的Codebook 背景建模算法. 实验表明, 本文算法不仅能够更准确地描述背景像素值在RGB 空间中的分布特征, 而且具有良好的鲁棒性.(The background modeling algorithm of mixture of Gaussian (MOG) and codebook is widely used in moving object detection of surveillance video. However, the ball model of MOG usually assumes that the three components of RGB are independent, while the cylinder model of codebook assumes that the value of background pixel is distributed uniformly within the cylinder and the changing direction of brightness points to the origin of the coordinate system. These assumptions reduce the descriptive capability for background modeling. Therefore, the paper proposes an ellipsoid- based background model, which overcomes the MOG and codebook0s limitations. By using principal component analysis to depict the ellipsoid background model, a novel PCA-based codebook background modeling algorithm is proposed. Experiments show that this algorithm can not only give more accurate description of the distribution of background pixels but also have a better robustness.)

用c#和emgucv实现了1.显示图像的直方图，2.直方图均衡化，3.中值滤波，4.sobel算子，roberts算子和Prewitt算子的边缘提取。各算法均自己实现不是调用函数。但是读图片的信息是用的emgucv。希望和同样刚接触emgucv的同学共同进步(With c# and emgucv achieve a 1 shows a histogram of the image, 2. Histogram equalization, 3 median filter, 4.sobel operator, edge roberts operator and Prewitt operator extraction. Various algorithms to achieve their own instead of calling a function. But reading the image information is used emgucv. The same hopes and common progress of students new to emgucv)

This is surf(Speeded Up Features Robust) matlab code for detection keypoint and construct descriptor and matching

读取图片，分割出敢兴趣区域，用到同态滤波，形态学，连通域，以及hough圆检测等算法。(Read images, segmented m interested in the area, the use of homomorphic filtering, connected domain, morphology, hough circle detection algorithm)

强跟踪滤波器，实现匀速目标和机动性目标的精确跟踪，达到了很好的跟踪效果(the case of tracking the moving target using the strong tracking filter)

说明：  下拉框自动提示 输入内容可直接显示出所匹配的所有信息(Drop-down box automatically prompt input all the information can be displayed directly matched)

引导滤波器，在引导图像的引导下对输入图像进行滤波，这种新型滤波器对边缘的保持很好，而且运行计算量和核的大小无关(Guide filters, under the guidance of the guide image filtering the input image, the edge of this new type of filter to keep well and run the calculation regardless of the size and nuclear)

说明：  基于小波变换的图像压缩和编码。matlab实现。(Based on wavelet transform image compression and coding. matlab achieve.)