北京大学遥感所的研究生图像处理讲义高级遥感数字图像处理（硕士生课程）之第二十一章 彩色和多光谱图像处理...

图片并附有说明，希望对大家有所帮助！-photo with a note, we want to help!

vrml与java3d结合的例子，用java3d加载vrml文件，包含了vrml97.jar和一些示列-VRML java3d combined with the examples used java3d loading VRML documents, and includes some show vrml97.jar out

数字识别代码 第一步：训练网络。使用训练样本进行训练。（此程序中也可以不训练，因为笔者已经将训练好的网络参数保存起来了，读者使用时可以直接识别）第二步：识别。 首先，打开图像（256色） 再次，进行归一化处理。点击“一次性处理” 最后，点击“R”或者使用菜单找到相应项来进行识别识别的结果显示在屏幕上，同时也输出到文件result.txt中 该系统的识别率一般为90％ 另外，也可以单独对打开的图片一步一步进行图像预处理工作。但要注意，每一步工作只能执行一遍，而且要按顺序执行。步骤为：“256色位图转为灰度图”“灰度图二值化”“去噪”“倾斜校正”“分割”“标准化尺寸”“紧缩重排” 另外注意，要识别的图片，要与win.dat、whi.dat位于同一目录。此两文件保存训练后网络的权值参数 -digital identification code first step : training network. Use of training samples for training. (These procedures also can not training because I have trained network parameters saved, Readers can direct identification) second step : identification. First, open the Image (256 colors) again, to a treatment. Click on the "one-stop" Finally, Click on the "R" or use the menu item corresponding to find the identification of the identification results on the screen, also result.txt output to a file which the system"s recognition rate of 90% for the other, can be individually opened to the step-by-step pictures for image prep

针对平行结构立体摄像机存在较大的遮挡边界区域问题，本文提出了一种基于可信图的遮挡 边界检测算法．首先通过采样计算立体图像对中几条扫描线的可信值，从而得到遮挡边界的若干采 样点；然后通过RANSAC算法得到遮挡边界．理论和实验分析表明该算法对复杂背景的立体图像有好 的检测精度；对人肩像等简单背景图像的检测精度不好，但通过校正后仍可得到较准确的遮挡边界．-Aimed at the existence of large occluded regions in stereo image pairs obtained from stereo camera pairs，an algorithm which can detect occluded boundary areas based on reliability map is proposed．Firstly，a number of sampled occluded boundary points are obtained by calculating reliability of the disparity on the corresponding sampled scan lines；secondly，the boundary line is calculated by RANSAC method．Theoretical and experimental analyses show that the proposed algorithm gives good boundary detection accuracies for stereo images with complex backgrounds． But for stereo images with homogenous background，results are not good，however the algorithm can also give relatively good accuracies after certain correction．

Borland C++ Builder 6.0开发的画图板程序，又一个寓教于乐的东东~-Borland C++ Builder 6.0 to develop the drawing board procedures, but also a recreation of the NEE ~

三维模型浏览器，可以浏览off模型，并实现了选取功能-Three-dimensional model of the browser, you can browse off model, and to achieve the selected function

一种快速的图像拐点的matlab程序，在一张图像上用9点快速检测，要求被处理图像为2值图像。-a fast image inflection point of Matlab procedures, in an image with 9:00 rapid detection, Image processing was required for two binary images.

多种二维图形的画法-variety of two-dimensional graphics of paint

人脸识别领域的K-L变换（Eigenface）的matlab源码-KL transform the field of face recognition (Eigenface) of matlab source

用deBoor_Cox算法生成的6次B样条曲线。-generation algorithm used deBoor_Cox six B-spline curves.