Alternative to abstract graph matching for locating objects from their salient features

摘要

The local-feature-focus method has become a standard means for robustly locating objects in two dimensions. Yet it is not without its difficulties, since the maximal clique approach to graph matching which it employs is excessively computation intensive, belonging to the class of NP-complete problems. Here we explore whether similar results can be obtained using other approaches, and in particular with the generalized Hough transform. The latter approach is found to be essentially equivalent to graph matching, while permitting objects to be located in polynomial (O(n4)) time.