Validation of polynomial-based equidistance fish-eye models

The majority of computer vision applications assume that the camera adheres to the pin-hole camera model. However, most real optical systems will introduce some undesirable effects, rendering the assumption of the pin-hole camera model invalid. By far the most evident of these effects is radial distortion, particularly in fish-eye camera systems where the level of this distortion is relatively extreme. The aim of fish-eye distortion correction is, therefore, to transform the distorted view of fish-eye cameras to the desired rectilinear pin-hole perspective view. To perform this distortion correction, several authors have developed models of fish-eye distortion. It is the aim of this paper to examine the accuracy of several of the polynomial-based models against the equidistance mapping function, which is the most common mapping function that fish-eye lenses are designed to follow.