New Image Mosaic Method Based on Fast Robust Correlation and Particle Filters
Jian Zhang · Information and Electronic Engineering · 2006
In order to mosaic images automatically without preprocessing and manual operations,a new approach that incorporates fast robust correlation with a particle filter to realize image mosaicking is presented in this paper.The parametric translation transformations found by fast robust correlation are used as predicted translation parameters of a particle filter.Using the predicted parameters,the particle filter can estimate image registration parameters easily and efficiently.Experimental results show that the proposed method can be robust,fast,exhaustive,and accurate to mosaic images.