Efficient one dimensional real scaled matching

Amihood Amir, Ayelet Butman, Moshe Lewenstein, Ely Porat, Dekel Tsur

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

Real Scaled Matching refers to the problem of finding all locations in the text where the pattern, proportionally enlarged according to an arbitrary real-sized scale, appears. Real scaled matching is an important problem that was originally inspired by Computer Vision. In this paper, we present a new, more precise and realistic, definition for one dimensional real scaled matching, and an efficient algorithm for solving this problem. For a text of length n and a pattern of length m, the algorithm runs in time O(n logm +√nm 3/2√logm).

Original languageEnglish
Pages (from-to)1-9
Number of pages9
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3246
DOIs
StatePublished - 1 Dec 2004
Externally publishedYes

Fingerprint

Dive into the research topics of 'Efficient one dimensional real scaled matching'. Together they form a unique fingerprint.

Cite this