Application of improved RANSAC algorithm to image registration
Chen Guilan · Journal of Computer Applications · 2010
In order to improve the speed of image registration,a fast method based on improved RANdom SAmple Consensus(RANSAC) algorithm was proposed.At first,Harris corner detector was used to extract the feature points in the reference and target images.Next,based on the proximity and similarity of their intensity,the feature points were matched.Finally,the improved RANSAC algorithm was used to estimate transform matrix more fast and accurately.To improve the speed of computation,a method of pre-detection was used to discard those temporary models that were not preview models.To delete outliers a random block selecting method was used to select samples,which improves the precisions.The experiment shows that this algorithm reduces the amount of computation largely and improves the speed of image registration when the precisions do not have notable change.