On Hausdorff Distance Measures
Shapiro, Matthew B. Blaschko · 2004
A number of Hausdorff-based algorithms have been proposed for finding objects in images. We evaluate different measures and argue that the Hausdorff Average distance measure outperforms other variants for model detection. This method has improved robustness properties with respect to noise. We discuss the algorithms with respect to typical classes of noise, and we illustrate their relative performances through an example edge-based matching task. We show that this method produces a maximum a posteriori estimate. Furthermore, we argue for improved computational efficiency by tree-like subdivisions of the model and transformation spaces.