Learning OpenCV

Calibration

Now that we have some idea of how we'd describe the intrinsic and distortion properties of a camera mathematically, the next question that naturally arises is how we can use OpenCV to compute the intrinsics matrix and the distortion vector. ^[172]

OpenCV provides several algorithms to help us compute these intrinsic parameters. The actual calibration is done via cvCalibrateCamera2(). In this routine, the method of calibration is to target the camera on a known structure that has many individual and identifiable points. By viewing this structure from a variety of angles, it is possible to then compute the (relative) location and orientation of the camera at the time of each image as well as the intrinsic parameters of the camera (see Figure 11-9 in the "Chessboards" section). In order to provide multiple views, we rotate and translate the object, so let's pause to learn a little more about rotation and translation.

Rotation Matrix and Translation Vector

For each image the camera takes of a particular object, we can describe the pose of the object relative to the camera coordinate system in terms of a rotation and a translation; see Figure 11-7.

Figure 11-7. Converting from object to camera coordinate systems: the point P on the object is seen as point p on the image plane; the point p is related to point P by applying a rotation matrix R and a translation vector t to P

In general, ...