A metric on the family of nonempty bounded subsets of a metric space defined as the maximum of the two directed suprema of distances from points of one set to the other (equivalently the infimal r such that each set is contained in the r-neighbourhood of the other).

Principle

Hausdorff distance organizes closeness of sets by the worst-case nearest-point gap: it measures how far one must travel from a point of either set to reach the other, symmetrizing directed one-sided proximities into a genuine metric on bounded closed sets.

Demonstration

In R with the Euclidean metric, the Hausdorff distance between [0,1] and [0.1,1.1] equals 0.1 because each point of one interval lies within 0.1 of some point of the other; small uniform perturbations of a compact set produce small Hausdorff distances.

Misapplication

Using Hausdorff distance for unbounded sets without modification (where it can be infinite) or expecting it to reflect measure-theoretic similarity (two sets can have small symmetric difference measure yet large Hausdorff distance) are common misuses.

Consequence

On the space of nonempty compact subsets of a compact metric space the Hausdorff distance defines a complete metric and makes the hyperspace compact; it is extensively used in shape analysis, geometric approximation, and convergence of sets.

Reversal

Replacing Hausdorff distance by symmetric difference measure or L^p set distances reverses the emphasis from geometric maximal pointwise separation to averaged or volumetric discrepancies; the reversal suits probabilistic or measure-theoretic comparisons but loses worst-case geometric control.

Boundary

Defined for nonempty bounded subsets of a metric space; often restricted to closed or compact sets to avoid pathologies and to ensure finiteness and metric properties; it depends on the ambient metric and is not intrinsic to the sets alone unless the ambient geometry is fixed.

Semantic Tension

Hausdorff distance competes with Gromov–Hausdorff distance and with measure-based set distances: the former compares metric spaces up to isometry, the latter averages discrepancies, so choice depends on whether pointwise geometric alignment or coarse volumetric similarity is desired.

Synthesis

Hausdorff distance is the maximal nearest-point gap between two bounded subsets: a symmetric, ambient-metric-dependent measure of set closeness that captures worst-case geometric deviation, underlies compactness results for hyperspaces, and contrasts with averaged set discrepancies.