少有人走的路GML是一款简单直接的相机标定软件，附上下载地址：http://graphics.cs.msu.ru/en/node/909。
下载后需要安装，在安装的目录下面有现成的棋盘格PDF文件，可以直接用于打印。我使用的是默认安装位置，其PDF文件位于C:\Program Files (x86)\GML\GML Camera Calibration 0.75\Calibration pattern目录下。
安装好之后，运行软件界面为：
选择File->New Project
弹出对话框如下，要求输入模板数量，这里模板数量是指你照片中棋盘格的种类数量，比如：只有照片中只有一张棋盘格就填写1，照片中有两张棋盘格就填写2，以此类推。
点击确认，会弹出对话框要求填写棋盘格参数：
再点击Add Image添加图片：
再点击Detect All检测照片中的棋盘格：
再点击Calibrate进行标定：
下面是勇哥摘抄的这个工具官网的说明档：
GML C++ Camera Calibration Toolbox
from: http://graphics.cs.msu.ru/en/node/909
Contact person: Alexander Velizhev (avelizhev@graphics.cs.msu.ru)
Downloads
Latest binaries:
Download GML C++ Camera Calibration Toolbox 0.75
Latest sources:Download GML C++ Camera Calibration sources 0.71
Download GML C++ Camera Calibration sources 0.4
Image samples:Download images
ChangeLog
16.07.2013 v0.75
A critical bug was fixed.
23.12.2011 v0.73
Two criticals bugs were fixed. Many thanks to Fedor Ermishin for the great feedback!
23.12.2011 v0.71
Some criticals bugs and memory leaks were fixed. Many thanks to Mathieu Peyréga for the great feedback! He helped us a lot to make this update.
18.11.2011 v0.7
Camera calibration using multiples patterns;
Support video data (usage of each N-frame);
Figure 1: GML Camera Calibration toolbox v. 0.7 supports multiple calibration patterns within one project. During the calibration a relative orientation between pattern coordinate systems is recovered. Our experiments show the usage of multiple patterns makes the calibration procedure more stable and impoves calibration accuracy. Also less images for calibration are needed.
.
06.02.2006 v0.4
Calibration module is OS independent;
Added image zoom functions;
Added a new tab "Point Density" for the points location control
27.10.2005 v0.31
Added a new function "Undistort"
Added an accuracy rating of parameters
Added new export formats
26.09.2005 v.0.1
First release
Introduction
GML Camera Calibration toolbox is a free functionally completed tool for cameras' calibrating. You can easy calculate intrinsic and extrinsic camera parameters after calibrating. This software uses a as a calibration pattern. The calibration process is completely automatic. There are two chessboard detection algorithms in all.
Figure 2: Toolbox interface.
Figure 3: Calibration results.
GML Camera Calibration toolbox 0.7 supports multiple calibration patterns within one project. During the calibration a relative orientation between pattern coordinate systems is recovered. The usage of multiple patterns makes the calibration procedure more stable, this impoves calibration accuracy and less images for calibration is needed.
Figure 4: Comparison of camera calibration using different number of patterns.
GML Camera Calibration toolbox 0.7 (beta) was released on 18.11.2011. Your Feedback is appreciated!
Requirements
The code was written for OpenCV 2.1 and was tested mostly on win32 (Windows XP, Windows Vista, Windows 7) platform. System Requirements: .Net Framework 3.5 SP1
Please read our recommendations for taking calibration photos before starting a calibration procedure.
Projects
Recommendations for taking calibration photos
GML MatLab Camera Calibration Toolbox
2005-2005
Team
Vladimir Vezhnevets, Lomonosov Moscow State University
Alexander Velizhev, Lomonosov Moscow State University
Anton Yakubenko, Lomonosov Moscow State University
Acknowledgements
We extend our gratitude to Chibunitchev A., whose help was significant in creation of this toolbox.
————————————————
版权声明：本文为CSDN博主「该-昵称已被占用」的原创文章，遵循CC 4.0 BY-SA版权协议，转载请附上原文出处链接及本声明。
原文链接：https://blog.csdn.net/u011091739/article/details/55258939
