Camera Calibration: Geometric and Colorimetric Correction
This chapter analyzes camera calibration from a geometric and colorimetric perspective. The first part of this chapter introduces the mathematical model that describes a camera (position, orientation, focal, etc.) as well as its applications (e.g. 3D reconstructions). This chapter also analyzes different geometric processes for stereoscopic images such as corrections of radial distortion as well as image rectification. The second part of this chapter will focus on colorimetric models relating to digital acquisition systems. We will demonstrate how to characterize a colorimetric camera and then analyze different elements involved in color correction for images taken from a system of cameras.
5.2. Camera calibration
The camera model used in computer vision is an idealized representation of a pinhole camera (see Chapter 1). This model, as well as problems such as projective geometry calibration, is analyzed by Hartley and Zisserman in [HAR 04], a key reference in the field of 3D vision as it is essential for all those interested in problems of this kind and the following sections are largely based on this work.
The objective of the matricial camera model is to represent this camera and its geometric relation to the scene. By extension, it also creates geometric relations between several cameras from which it is possible to create 3D reconstructions of the scene or locate an object. ...