HUinvmoments

基于CFAR虚警率的图像分割算法，可快速对灰度图像进行CFAR检测分割。(Based on the false alarm rate of CFAR image segmentation algorithm, the gray image can be fast CFAR detection and segmentation.)

本程序用于对工件图像的预处理，边缘提取等。将改进遗传算法和Hausdorff距离结合用于工件识别。采用canny边缘为匹配特征，将修正的Hausdorff距离作为相似性度量准则，并应用遗传算法进行最佳匹配的快速搜索，成功实现了目标物体的匹配识别(This procedure used for the workpiece image preprocessing, edge extraction and so on. Will improve the genetic algorithm and Hausdorff distance combination for workpiece identification. Using canny edge for matching characteristics, will be amended Hausdorff distance as similarity measurement criteria and the application of genetic algorithm is the best match of the Quick Search, the successful achievement of the target object matching identification)

初学者的游程编码，可以进行小图像的压缩，可以以此为基础进行编码和参考(RLE beginner, you can make small image compression, can be encoded as a basis and reference)

基于分水岭自动选取种子的区域生长来进行彩色图像分割(Automatically selected based on the watershed to the seed region growing for color image segmentation)

计算图像的多重分形谱。它是分形的多尺度衡量的分型参数。(Calculated images multifractal spectrum. It is a multi-fractal measure the sub-type parameters.)

基于visual c++平台，相机标定方法中的直接线性变换方法，即DLT方法(Based on visual c++ platform, camera calibration method of direct linear transformation method, namely DLT method)

基于GA和MP的信号稀疏分解,算法速度比元来的有很多改进。(Based on GA and MP sparse signal decomposition, the algorithm faster than the yuan to have many improvements.)

有关图像融合的一个比较全的代码集合，有gui界面，可选择不同的处理图像方法融合图像(the image of a more integrated whole code set, a gui interface choice of different methods of image processing Image Fusion)

可控金字塔，用于图像去噪，识别，是一种方向可控的金字塔结构的图像处理方法(steerable pyramids for image denoising, recognition, is a steerable pyramid image processing method)

说明：  实现直接序列扩频水印的嵌入与提取。基于待嵌入消息生成二进制水印序列,生成扩频码，用消息序列调制生成水印序列。依次处理灰度图像的每一个载体图像像素，根据相应位置 水印序列完成嵌入修改。(Realize the embedding and extraction of direct sequence spread spectrum watermark. The binary watermark sequence is generated based on the message to be embedded, the spread spectrum code is generated, and the watermark sequence is generated by message sequence modulation. Each carrier image pixel of gray image is processed in turn, and the embedding modification is completed according to the corresponding position watermark sequence.)