threshold-roc

将黑白图像转换成灰度级—邻域平均灰度级二维直方图(Black and white images will be converted into gray-scale- the average gray-scale neighborhood of two-dimensional histogram)

基于全局直方图均衡图像增强和暗通道优先理论相结合的去雾方法。(The priority theory based on global histogram equalization image enhancement and dark channels combined to fog.)

NSCT分解 图像融合 各个融合指标评价体系 他是分解(NSCT decomposition image fusion)

SAR图像的非相干cloude分解，最后得到熵值、散射角、反熵。合成三色彩色图，最终将结果显示出来。(SAR image non coherent cloude decomposition , finally entropy,scatter degree were obtaind.)

说明：  实现机器视觉双目标定，三维重建等算法，具有很好地擦考价值( Robot with two cameras to switch recognition system. (Note that not binocular vision) and the reference used opencv1.0 cmv. )

可以设置图像分块的大小和分块数，然后来分块保存图像到文件目录下面(You can set the size of the image sub-block and sub-blocks, and then to block to save the image to a file directory)

zemax非序列下gaussian光束的matlab程序（包括zemax演示源文件）(zemax sequence under non-gaussian beam matlab program (including zemax demo source files))

基于JPEG压缩的无参考图像质量评价算法，输入一幅图像，输出图像的评价值，值在0-100之间，越接近0图像的质量越好(JPEG-based compression algorithm for no-reference image quality assessment, an input image, the output image uation value, a value between 0-100, closer to 0 the better the image quality)

用lbp实现纹理特征提取，并分类说明: 一共有三个m文件，一个是lbp.m, 存放主要的lbp算法， 一个是getmapping,用以做算法的辅助函数， 一个是lbptest.m,存放着测试代码。 这三个文件需要放到同一个文件夹，并在文件夹中添加相应的图片， 具体的图片名字见lbptest.m的代码，运行lbptest.m可以查看结果。代码最后给出效果图 这三个文件是最传统的LBP方法，有256种。(With lbp achieve texture feature extraction and classification Description: A total of three m files, one is lbp.m, store the main lbp algorithm, One is getmapping, helper functions for doing arithmetic, One is lbptest.m, kept the test code. These three documents need to put the same folder, and add the appropriate folder, pictures, See lbptest.m specific name of the picture of the code, run lbptest.m can view the results. Finally renderings code These three documents are the most traditional LBP method, there are 256 kinds.)

YUV多种图像处理方式，包括yuv2rgb,rgb2yuv.等等。但需要仔细调试(YUV multiple image processing methods, including yuv2rgb, rgb2yuv. And so on. But needs to carefully debugging)