Geometric pattern matching in d-dimensional space

L. P. Chew, D. Dor, Alon Efrat, K. Kedem

Research output: Contribution to journalArticle

27 Citations (Scopus)

Abstract

We show that, using the L∞ metric, the minimum Hausdorff distance under translation between two point sets of cardinality n in d-dimensional space can be computed in time O(n(4d-2)/3 log2 n) for 3 < d ≤ 8, and in time O(n5d/4 log2 n) for any d > 8. Thus we improve the previous time bound of O(n2d-2 log2 n) due to Chew and Kedem. For d = 3 we obtain a better result of O(n3 log2 n) time by exploiting the fact that the union of n axis-parallel unit cubes can be decomposed into O(n) disjoint axis-parallel boxes. We prove that the number of different translations that achieve the minimum Hausdorff distance in d-space is Θ(n[3d/2]). Furthermore, we present an algorithm which computes the minimum Hausdorff distance under the L2 metric in d-space in time O(n[3d/2]+1+δ), for any δ > 0.

Original languageEnglish (US)
Pages (from-to)257-274
Number of pages18
JournalDiscrete and Computational Geometry
Volume21
Issue number2
StatePublished - 1999
Externally publishedYes

Fingerprint

Pattern matching
Pattern Matching
Hausdorff Distance
Minimum Distance
D-space
Metric
Unit cube
Set of points
Cardinality
Disjoint
Union

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computational Theory and Mathematics
  • Discrete Mathematics and Combinatorics
  • Geometry and Topology

Cite this

Geometric pattern matching in d-dimensional space. / Chew, L. P.; Dor, D.; Efrat, Alon; Kedem, K.

In: Discrete and Computational Geometry, Vol. 21, No. 2, 1999, p. 257-274.

Research output: Contribution to journalArticle

Chew, LP, Dor, D, Efrat, A & Kedem, K 1999, 'Geometric pattern matching in d-dimensional space', Discrete and Computational Geometry, vol. 21, no. 2, pp. 257-274.
Chew, L. P. ; Dor, D. ; Efrat, Alon ; Kedem, K. / Geometric pattern matching in d-dimensional space. In: Discrete and Computational Geometry. 1999 ; Vol. 21, No. 2. pp. 257-274.
@article{d8367ccd07364d6c8edf74d1ef07761e,
title = "Geometric pattern matching in d-dimensional space",
abstract = "We show that, using the L∞ metric, the minimum Hausdorff distance under translation between two point sets of cardinality n in d-dimensional space can be computed in time O(n(4d-2)/3 log2 n) for 3 < d ≤ 8, and in time O(n5d/4 log2 n) for any d > 8. Thus we improve the previous time bound of O(n2d-2 log2 n) due to Chew and Kedem. For d = 3 we obtain a better result of O(n3 log2 n) time by exploiting the fact that the union of n axis-parallel unit cubes can be decomposed into O(n) disjoint axis-parallel boxes. We prove that the number of different translations that achieve the minimum Hausdorff distance in d-space is Θ(n[3d/2]). Furthermore, we present an algorithm which computes the minimum Hausdorff distance under the L2 metric in d-space in time O(n[3d/2]+1+δ), for any δ > 0.",
author = "Chew, {L. P.} and D. Dor and Alon Efrat and K. Kedem",
year = "1999",
language = "English (US)",
volume = "21",
pages = "257--274",
journal = "Discrete and Computational Geometry",
issn = "0179-5376",
publisher = "Springer New York",
number = "2",

}

TY - JOUR

T1 - Geometric pattern matching in d-dimensional space

AU - Chew, L. P.

AU - Dor, D.

AU - Efrat, Alon

AU - Kedem, K.

PY - 1999

Y1 - 1999

N2 - We show that, using the L∞ metric, the minimum Hausdorff distance under translation between two point sets of cardinality n in d-dimensional space can be computed in time O(n(4d-2)/3 log2 n) for 3 < d ≤ 8, and in time O(n5d/4 log2 n) for any d > 8. Thus we improve the previous time bound of O(n2d-2 log2 n) due to Chew and Kedem. For d = 3 we obtain a better result of O(n3 log2 n) time by exploiting the fact that the union of n axis-parallel unit cubes can be decomposed into O(n) disjoint axis-parallel boxes. We prove that the number of different translations that achieve the minimum Hausdorff distance in d-space is Θ(n[3d/2]). Furthermore, we present an algorithm which computes the minimum Hausdorff distance under the L2 metric in d-space in time O(n[3d/2]+1+δ), for any δ > 0.

AB - We show that, using the L∞ metric, the minimum Hausdorff distance under translation between two point sets of cardinality n in d-dimensional space can be computed in time O(n(4d-2)/3 log2 n) for 3 < d ≤ 8, and in time O(n5d/4 log2 n) for any d > 8. Thus we improve the previous time bound of O(n2d-2 log2 n) due to Chew and Kedem. For d = 3 we obtain a better result of O(n3 log2 n) time by exploiting the fact that the union of n axis-parallel unit cubes can be decomposed into O(n) disjoint axis-parallel boxes. We prove that the number of different translations that achieve the minimum Hausdorff distance in d-space is Θ(n[3d/2]). Furthermore, we present an algorithm which computes the minimum Hausdorff distance under the L2 metric in d-space in time O(n[3d/2]+1+δ), for any δ > 0.

UR - http://www.scopus.com/inward/record.url?scp=0033478042&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033478042&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0033478042

VL - 21

SP - 257

EP - 274

JO - Discrete and Computational Geometry

JF - Discrete and Computational Geometry

SN - 0179-5376

IS - 2

ER -