A novel warping distance measure

In this paper we present a novel distance measure, the minimum landscape distance (MLD). MLD provides a non-linear mapping between the elements in one sequence to those of another. Each element in one sequence is mapped to that with the highest neighbourhood structural similarity (landscape) in the other sequence within a search window. Experimental results obtained on sequences representing binary images show that MLD is superior to the Euclidean, correlation, and dynamic time warping (DTW) distance measures. We experimented with various sequence, landscape, and search window sizes.