Image Registration by Sequential Tests of Hypotheses: Gaussian and Binomial Techniques

摘要

The problem of translational image registration has received considerable attention in the area of image processing and recognition as applied to remote sensing. The main methods that have been proposed are based either on correlation techniques or on algorithms of the Type SSDA (Sequential Similarity Detection Algorithm), where the error between the two images is accumulated and a threshold sequence is selected, such that the rejection of a candidate position can be done at an early stage. A new approach to image registration problems is proposed, based on the theory of sequential test of hypotheses. This leads to the development of two different methods: the first one is based on the Gaussian assumption and uses the fact that the variance of the error between two images to be registered tend to be low on the registration point. The second uses binary images derived from the original ones. The statistical model for the resulting accumulated error is a binomial distribution and the registration position is characterized by a low probability of the binary error being one. In both methods two sequences of thresholds are employed: one leading to the rejection of the point and the other one to the eventual acceptance of it. Experimental results with both methods are presented.