Pearce, A. R., Caelli, T., and Bischof, W. F. (1993). Rulegraphs for pattern recognition. Proceedings of the First Asian Computer Vision Conference, Osaka, Japan, Nov 23-25, pp. 79-82.
We present and compare two new techniques for learning Relational Structures (RS) as they occur in 2D pattern adn 3D object recognition. The two techniques, Evidence-based Systems (EBS) and Rulegraphs (RG) combine techniques from Computer Vision with those from Machine Learning, Uncertainty Theory and Graph Matching. EBS have the havility to generalize pattern rules from training instances in terms of bounds on both unary (singe part) and binary (part relation) numerical features. Rulegraphs check the compatibility between rules by combining Evidence Theory with Graph Matching techniques. The two systems are tested and compared using a difficult pattern recognition problem.
Back to publications.