xbchuzao

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)

实验证明，本方案对于帧间位移较为明显的单个平移运动刚体，特别是外部轮廓较为平滑的物体具有良好的分割与跟踪效果。(experiment proved that the program for the inter-frame displacement is quite obvious that the single translation rigid movement, particularly smooth external contours of objects more good segmentation and tracking results.)

说明：  matlab编写的纹理图像分割 gussian滤波后k-means聚类 并将不同区域用线条表示出来(Matlab prepared texture image segmentation, Gussian filtering, K-means clustering and different regions are represented by lines)

对立体视觉的双目图像进行极线校正，以实现双目致密匹配(Of three-dimensional visual image of the binocular epipolar rectification to realize dense binocular matching)

该程序主要是对一幅图像进行动漫风格化 总体上，把照片变成动漫化风格分以下几步： （1）膨胀图片让其亮暗分明 （2）区域平滑 （3）增强锐化，以突出显示边缘线条。(The program is mainly a cartoon style animation style On the whole, take the photo into the animation style and divide the following steps. (1) the expansive picture makes it bright and dark (2) regional smoothness (3) sharpening to highlight the edge lines.)

说明：  针对现有方法存在的隐藏容量低、操作不灵活的缺陷，文章提出了一种基于无损压缩的加密图像可逆 信息隐藏算法。(In view of the defects of low hiding capacity and inflexible operation of the existing methods, this paper proposes a reversible information hiding algorithm for encrypted images based on lossless compression.)

softcast的2d dct实现，先对图像转灰度图，得到的亮度信号矩阵做DCT 变换(softcast of 2d dct implementation, the first image to grayscale, luminance signal obtained by DCT transformation matrix do)

jpeg图像压缩和解压缩过程，使用的哈弗曼表包含于程序中，内有详细的文件说明(jpeg image compression and decompression process, the use of Huffman tables included in the program, with detailed documentation)

道路识别的一个matlab仿真程序，主要是基于道路颜色的图像处理。(A road recognition matlab simulation program was based primarily on the road color image processing.)

说明：  通过matlab，可以快速得到图像的信噪比，峰值信噪比等。里面也有一些解释(Through MATLAB, the signal-to-noise ratio and peak signal-to-noise ratio of image can be obtained quickly. There are also some explanations)