Structural Matching in Computer Vision Using Probabilistic Relaxation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Triangle refinement in a constrained Delaunay triangulation skeleton
Pattern Recognition
| Hi-index | 0.00 |
Graph matching algorithms have proven to be useful in many applications, such as character recognition, shape analysis and image analysis. As the graph matching problem is one of exponential computational complexity, various heuristics and estimations have been proposed. While many of these algorithms succeed in improving the time taken to perform a match, most do not guarantee that an optimal solution will be found. It is the aim of the proposed algorithm to reduce the complexity of the graph matching process, while still producing an optimal solution for a known application. This is achieved by removing a graph edit operation from the matching process, and compensating for the lost robustness by introducing a hierarchical matching process that is centered around an application-specific criterion that operates on the subgraph scale. Results show that the proposed algorithm is faster than two previous methods that are based on graph edit operations.