Shift detection by restoration-demonstrated by signal based pointpattern matching

摘要

Shift detection is one of the fundamental tasks within the fieldof picture registration. With SDR (shift detection by restoration) arobust signal-based procedure is given for detection of real shifts ofpictures. On the basis of a model which contains various perturbances,the shifted delta impulse determining the shift is restored with methodsof image restoration. By the use of the magnitude spectrum of twopictures also rotation and scaling can be detected by SDR. After a briefintroduction to the basics of SDR, the procedure is demonstrated in afurther problem-point pattern matching (PPM). A new extremely robustmethod is derived whereby pictures are generated from point patterns,which are used for the matching. An evaluation of the restored deltaimpulse is used, in order to determine weights for point correspondences