A fast and robust image registration method based on an early consensus paradigm
Pattern Recognition Letters
Hi-index | 0.00 |
This paper presents a fast and extremely robust feature-basedmethod for planar registration of partly overlappingimages that uses a two-stage robust fitting approach comprisinga fast estimation of a transformation hypothesis(that we show is highly likely to be correct) followed by aconfirmation and refinement stage. The method is particularlysuited for automatic stitching of oversize documentsscanned in two or more parts. We show simulations, alsosupported by practical experiments, that prove both the robustnessand computational efficiency of the approach.