Research on algorithm of image matching based on distance of information-state transition

Abstract

Image matching is considered as an important technology on Digital Image Processing. The goal of image matching is to find out the spatial relationship between two different images. DIT is a kind of measurement method which describes the obstacles during the information-state transition. Research methods of DIT can exactly reflect the quantificational meaning for the qualitative description of the information-state transition. This paper introduces the calculating method of DIT from Shannon entropy and the concept of DIT into image matching. Finally, inspired by EMD, an original image-matching algorithm based on DIT is proposed, which is from the EMD method.