小波分析

压缩感知CS——采用小波变换进行稀疏表示，高斯随机矩阵为观测矩阵，重构算法为SP算法，对256*256的lena图处理，比较原图和SP算法在不同采样比例(0.74、0.5、0.3)下的重构效果，并各运行50次，比较算法性能PSNR和每次的运行时间(Compressed sensing CS- using wavelet transform as sparse representation, Gaussian random matrix as the observation matrix and SP algorithm as the reconstruction algorithm. Compare lena figure and the reconstruction results using SP algorithm at different sampling ratio (0.74,0.50.3),then each runs 50 times, compare the performance of PSNR and each running time)

VC做的图像处理方面的一些代码，仅供参考(VC do image processing of some code, only supplies the reference )

图像处理方面，用VC++实现的连通区域标记的程序。(Image processing, with VC++ implementation of the procedures connected region markers.)

针对大尺度图像配准和不同传感器图像配准问题, 介绍了一种基于 SIFT 的图像配准方 法。(For large-scale image registration and image registration problem from different sensors, introduces a SIFT-based image registration method.)

常规的中值滤波器，在噪声的密度不是很大的情况下效果不错。但是当概率出现的概率较高时，常规的中值滤波处理后，仍然具有噪声点，并丢失了细节和边缘，效果不是很好。 自适应中值滤波器可以尽可能的保护图像中细节信息，避免图像边缘的细化或者粗化。(The conventional median filter works well when the noise density is not very large. However, when the probability of probability is high, the conventional median filter still has noise points, and the details and edges are missing. The result is not very good. The adaptive median filter can protect details in the image as far as possible and avoid thinning or coarsening of the image edges.)

说明：  前馈神经网络图像去噪，DnCNN的MATLB实现，以及相应的峰值信噪比和结构相似性的计算。(DnCNN,MATLAB,And calculation of peak signal-to-noise ratio structural similarity)

winform 自定义窗体、控件

医学图像处理：P0701：细胞边缘检测；702：癌细胞形态学分析；P0703：癌细胞颜色分析(Medical Image Processing: P0701: cell edge detection 702: morphological analysis of cancer cells P0703: color analysis of cancer cells)

说明：  小波模极大值用于边缘特征提取.很好的一个matlab算法，可以直接验证。(Wavelet modulus maxima for edge feature extraction algorithm matlab good one can be directly verified.)

广度优先搜索和深度优先搜索在网络拓扑图上的实现。广度优先搜索是从某一节点开始，搜索与其线连接的所有节点，按照广度方向像外扩展，直到不重复遍历所有节点；深度优先搜索是从某一节点开始，沿着其搜索到的第一个节点不断深入下去，当无法再深入的时候，回溯节点，然后再在回溯中的某一节点开始沿另一个方向深度搜索，直到不重复的遍历所有节点。 (Breadth-first search and depth-first search in the network topology is realized. Breadth-first search is started from a node, searching all the nodes connected to its line, in accordance with the breadth direction as extracapsular extension, until no repeat traverse all nodes depth-first search is started from a node along its search to the first nodes deepening it, when it can not go any further, backtracking node, and then back again in the beginning of a node depth search in the other direction until the non-repetition of the traverse all nodes.)