<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[mppcasc的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/mppcasc</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; mppcasc]]></copyright><item><title><![CDATA[乌班图安装 Kalibr]]></title><link>https://blog.youkuaiyun.com/mppcasc/article/details/120239157</link><guid>https://blog.youkuaiyun.com/mppcasc/article/details/120239157</guid><author>mppcasc</author><pubDate>Sat, 11 Sep 2021 20:56:42 +0800</pubDate><description><![CDATA[安装ROS Melodic
1 Installation
1.1 Configure your Ubuntu repositories
http://www.360doc.com/content/18/0417/15/54525756_746389833.shtml
2.2 Setup your sources.list
sudo sh -c 'echo "deb http://packages.ros.org/ros/ubuntu $(lsb_release -sc) main" &gt; /etc/ap]]></description><category></category></item><item><title><![CDATA[OpenCV之getOptimalNewCameraMatrix]]></title><link>https://blog.youkuaiyun.com/mppcasc/article/details/120004012</link><guid>https://blog.youkuaiyun.com/mppcasc/article/details/120004012</guid><author>mppcasc</author><pubDate>Tue, 31 Aug 2021 20:23:14 +0800</pubDate><description><![CDATA[getOptimalNewCameraMatrix
在内部调用cvGetOptimalNewCameraMatrix()。
主要步骤如下：

首先获得内切和外切矩形
根据上述内切和外切矩形，分别计算新的投影矩阵
根据alpha，由上述两个投影矩阵插值得出最终的投影矩阵

获取内切和外切矩形
Get inscribed and circumscribed rectangles in normalized (independent of camera matrix) coordinates.
通过将整个图像划分]]></description><category></category></item><item><title><![CDATA[基于2D特征的运动估计]]></title><link>https://blog.youkuaiyun.com/mppcasc/article/details/119811049</link><guid>https://blog.youkuaiyun.com/mppcasc/article/details/119811049</guid><author>mppcasc</author><pubDate>Sun, 29 Aug 2021 17:20:35 +0800</pubDate><description><![CDATA[estimateGlobalMotionLeastSquares
基于最小二乘的全局运动估计

]]></description><category></category></item><item><title><![CDATA[RANSAC算法]]></title><link>https://blog.youkuaiyun.com/mppcasc/article/details/119766230</link><guid>https://blog.youkuaiyun.com/mppcasc/article/details/119766230</guid><author>mppcasc</author><pubDate>Sun, 29 Aug 2021 13:48:28 +0800</pubDate><description><![CDATA[RANSAC算法

]]></description><category></category></item><item><title><![CDATA[畸变模型的求解]]></title><link>https://blog.youkuaiyun.com/mppcasc/article/details/119766074</link><guid>https://blog.youkuaiyun.com/mppcasc/article/details/119766074</guid><author>mppcasc</author><pubDate>Thu, 19 Aug 2021 20:12:16 +0800</pubDate><description><![CDATA[畸变模型的求解
opencv之undistortPoints

]]></description><category></category></item><item><title><![CDATA[图像畸变处理]]></title><link>https://blog.youkuaiyun.com/mppcasc/article/details/119765418</link><guid>https://blog.youkuaiyun.com/mppcasc/article/details/119765418</guid><author>mppcasc</author><pubDate>Wed, 18 Aug 2021 19:59:14 +0800</pubDate><description><![CDATA[图像畸变
In geometric optics, distortion is a deviation from rectilinear projection; a projection in which straight lines in a scene remain straight in an image. It is a form of optical aberration.
畸变是像差的一种，它在感官上给人一种错觉（即原本直线变成了曲线），但图像的信息不会丢失。
本文主要介绍畸变的两种处理方式，一]]></description><category></category></item><item><title><![CDATA[LM优化算法]]></title><link>https://blog.youkuaiyun.com/mppcasc/article/details/119491223</link><guid>https://blog.youkuaiyun.com/mppcasc/article/details/119491223</guid><author>mppcasc</author><pubDate>Sat, 07 Aug 2021 19:19:00 +0800</pubDate><description><![CDATA[LM算法理论知识梯度下降高斯牛顿Levenberg–Marquardt算法框架算法的整体流程求解器update流程说明算法实现头文件cpp算法调用
LM优化算法，是一种非线性优化算法，其可以看作是梯度下降和高斯牛顿法的结合。综合了梯度下降对初值不敏感和高斯牛顿在最优值附近收敛速度快的特点。
本人非数学专业，且对算法理解可能不到位，详细的算法推导及各个优化算法之间的关系，非常推荐看**《METHODS FOR NON-LINEAR LEAST SQUARES PROBLEMS 》**，其介绍更详细也更专业。
]]></description><category></category></item></channel></rss>