Camera Calibration for Video Surveillance Toolbox

A matlab toolbox for easy camera calibration in video surveillance.

Introduction
Most cameras for video surveillance are installed in high positions with a slightly tilted angle. In such installation, it is possible to retain only three calibration parameters in the original camera model, namely the focal length (f), tilting angle (theta) and camera height (c). These parameters can be directly solved by a nonlinear curve fitting using the observed head and foot points of a walking human. This toolbox privdes several functions and sample data needed to demonstrate this method.

List of scripts (demo)

Calibrating camera using head/foot points data

CalibrationDemo.m
Evaluating performance of calibration and height estimation

CrossValidationY.m - using Y only
CrossValidationYd.m - using Y only (distortion corrected)
CrossValidationXY.m - using X and Y
CrossValidationXYd.m - using X and Y (distortion corrected)
Demonstrating distortion correction

UndistortionDemo.m
List of functions

For given foot points, compute head points

FootToHeadY.m - using Y only
FootToHeadXY.m - using X and Y
FootToHeadXYd.m - using X and Y (distortion corrected)
For given foot and head points, compute human height

PointsToHeightY.m
Distort or undistort image

DistortImage.m
UndistortImage.m
Distort or undistort points

DistortPoints.m
UndistortPoints.m
Sample data

Sample data is under /data folder.

Homepage

http://vision.inha.ac.kr

License

This code is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License version 2 as published by the Free Software Foundation.

(C) 2015 Shengzhe Li [e-mail]