2024 Volume 42 Issue 8 Pages 793-796
In this paper, we propose a method to remove distortion from highly-distorted input images by an ultra-wide-angle vision sensor. We extend an extensive look-up table step by step by taking into account reduction of chromatic aberration, as iteratively concatenating look-up tables for distortion removal using projective transformation of a local small square lattice pattern. Comparative evaluation experiments of RGB values before and after removing distortion were conducted. Results by our proposed method demonstrated effectiveness for more reliable feature point matching and object recognition by reducing chromatic aberration.