Internal Camera Calibration using Rotation and Geometric Shapes

Abstract

This paper describes a simple method for internal camera calibration for computer vision systems. It is intended for use with medium to wide angle camera lenses. With modification it can be used for longer focal lengths. This method is based on tracking image features through a sequence of images while the camera undergoes pure rotation. This method does not require a special calibration object. The location of the features relative to the camera or to each other need not be known. It is only required that the features can be located accurately in the image. This method can therefore be used both for laboratory calibration and for self calibration in autonomous robots working in unstructured environments. The method works when features can be located to single pixel accuracy, but subpixel accuracy should be used if available. In the basic method the camera is mounted on a rotary stage so that the angle of rotation can be measured accurately and the axis of rotation is constant. A set o...