MRF-grubcut

说明：  对图像进行小波分解后，并抽取出分解后的垂直高频分量，并将其保存出来。(Image after wavelet decomposition, and to take out after the decomposition of the vertical high-frequency component, and save it out.)

说明：  直方图均衡化能自动地增强整个图像的对比度，但它的具体增强效果不好控制，处理的结果总是得到全局均衡化的直方图，对数据的处理不加选择，可考虑采用自适应直方图均衡化（AHE），可改进图像的局部对比度以及获取更多的图像细节。MATLAB代码均可直接运行实现。(Digital image processing histogram equalization,Adaptive histogram equalization)

ONSL0稀疏重构算法是NSL0的优化版 输入: y:测量值向量 A:测量矩阵 A_pinv:A的广义逆 输出: xr:重构信号 用于对信号或者图像的压缩重构(ONSL0 u7A00 u758F u91CD u6784 u7B97 u6CD5 u662FNSL0 u7684 u4F18 u5316 u7248 u8F93 u5165: y: u6D4B u91CF u503C u5411 u91CF A: u6D4B u91CF u77E9 u9635 A_pinv: A u7684 u5E7F u4E49 u9006 u8F93 u51FA: xr: u91CD u6784 u4FE1 u53F7 u7528 u4E8E u5BF9 u4FE1 u53F7 u6216 u8005 u56FE u50CF u7684 u538B u7F29 u91CD u6784)

asyDraw是一个简单的绘图软件，这个实例会综合运用本章所介绍了Visual C++ 2005的基础知识(this program for paint)

这是我的大作业，作用是测试车流量并计数，使用高斯模型对车辆进行定位分割，已验证成功内附测试视频(This is my big job, the role is to test the traffic flow and counting, the use of the Gauss model of the vehicle positioning and segmentation, has been verified successfully included test video)

运用openCV编写的C++图像色彩转移程序，原理是先转化到Lab图像，再保持L不变调整图像的a、b分量的均值跟方差。(Using openCV prepared C++ image color transfer process, the principle is first converted to Lab image, and then adjust the image to keep L constant a, b, with the mean variance component.)

改程序是用来行人检测的，有源代码，可以轻松学习并参考。(The reform program is used for pedestrian detection, source code, you can easily learn and reference.)

说明：  多阈值分割 ，基于otsu生成阈值，有用的可以试试(Multi threshold segmentation, threshold generation based on Otsu, useful can try)

说明：  一个Excel二次开发SDK接口类附代码。(An Excel secondary development SDK interface class with code)

本设计主要目的是实现染色体的识别计数。首先对图形进行预处理，将图像转化为灰度图、滤波去噪，然后对图像进行二值化，合理选取阈值，进行直方图阈值处理，最后去除图像中面积过小的杂点，最终实现识别计数。(The main purpose of this design is to achieve the identification of chromosome count. The graphical pre-processing, the image is converted to grayscale, filtering denoising, image binarization, reasonable selection of the threshold, histogram thresholding, and finally remove the image area is too small noise , and ultimately identify the count.)