算法

里面包含kalman，扩展kalman，无迹kalman，粒子滤波，无迹粒子滤波等源码的实现。(upf--The Unscented Particle Filter )

matlab编写的截断半径45时，低通滤波器对与图像的影响，画图比较滤波前后的对比图。(matlab truncation radius of the preparation of 45, the low-pass filter and image of the effect of drawing the contrast to compare before and after filtering Fig.)

用粒子滤波器实现的多运动员跟踪，包含了data文件(Using Particle Filters athletes realize the multi-tracking, contains a data file)

机器人的轨迹跟踪程序，修改参数可以重新建立自己想要的模型，可以看到末端轨迹的运动画面。(Robot trajectory tracking program, you can modify the parameters to re-establish the model you want, you can see the end of the trajectory of the moving picture.)

小波脊线提取crazy climber算法的文献出处(The wavelet ridge extract crazy climber algorithm literature source)

基于阴影形成的原理及阴影区域的特点，研究阴影区域的特征表示方法； 基于阴影区域特征设计或改进阴影检测方法，检测出户外图像中的阴影区域； 在检测出阴影区域的基础上，研究阴影去除方法； 编程实现相关方法并进行实验验证及分析。(Shadow detection and removal of single image)

能够实现图像的纹理检测，可以直接调入一个图像然后得到纹理(to achieve the texture image detection, can be directly transferred to a texture image is then)

何凯明的基于暗原色先验的去雾代码，去雾效果很好(Kaiming dark colors based on a priori defogging code defogging good effect)

根据icip会议2005年度的一篇文章编写的matlab程序。包含有效果图。 运用smqt（连续均值量化变换）算法进行灰度图像增强。它比传统方法（直方图均衡化处理）的优点在于可以保证增强后图象的直方图波形不变，可以展现图象的底层结构，自动集中改进图象细节。是一种新的效果更好的灰度图象增强算法。(under icip 2005 annual meeting of an article prepared by the Matlab procedures. Includes map results. Smqt use (continuous quantitative mean transform) algorithm for gray-scale image enhancement. It is superior to the traditional method (histogram equalization processing) has the advantage can guarantee enhanced image histogram waveform unchanged. able to display images of the bottom structure, automatic focus on improving image details. Is a new and better results gray image enhancement algorithms.)

这些代码主要是IDL语言源码，主要功能：初始化ENVI,定标,大气校正,融合,镶嵌,裁剪,分类(监督+非监督),面向对象特征提取,结束ENVI(These codes are mainly IDL language source code, the main functions: initialization, ENVI, calibration, atmospheric correction, fusion, mosaic, cutting, classification (supervision + non supervision), object oriented feature extraction, ending ENVI)