An envelope-based approach to rotation-invariant boundary image matching

초록

Supporting the rotation invariance is crucial to provide more intuitive matching results in boundary image matching. Computing the rotation-invariant distance, however, is a very time-consuming process since it requires a lot of Euclidean distance computations for all possible rotations. To solve this problem, we propose a novel notion of envelope-based lower bound, and using the lower bound we reduce the number of distance computations dramatically. We first present a single envelope approach that constructs a single envelope from a query sequence and obtains a lower bound of the rotation-invariant distance using the envelope. This single envelope approach, however, may cause bad performance since it may incur a smaller lower bound due to considering all possible rotated sequences in a single envelope. To solve this problem, we present a concept of rotation interval, and using it we generalize the single envelope lower bound to the multi-envelope lower bound. Experimental results show that our envelope-based solutions outperform existing solutions by one to three orders of magnitude. © 2011 Springer-Verlag.