Point pattern matching by a genetic algorithm

Nirwan Ansari, Ming Hwang Chen, Edwin S.H. Hou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

15 Scopus citations

Abstract

The problem of finding a subset of points in a pattern that best match to a subset of points in another pattern through a transformation in an optimal sense is considered. An exhaustive search to find the best assignment mapping one set of points to another set is, if the number of points that are to be matched is large, computationally expensive. A genetic algorithm that searches for the best ('almost the best') assignment efficiently is described. To map the point pattern matching into the framework of a genetic algorithm, a fitness function that is inversely proportional to the match error and a scheme for encoding an assignment between two sets of points into a string are used, along with a genetic operator known as the mixed-type partial matching crossover. Experimental results have demonstrated the robustness and the fast convergence of the algorithm. The algorithm can be applied to n-dimensional point patterns and any transformation. Results are presented for two-dimensional point patterns and a similarity transformation.

Original languageEnglish (US)
Title of host publicationPower Electronics Emerging Technologies
PublisherPubl by IEEE
Pages1233-1238
Number of pages6
ISBN (Print)0879426004
StatePublished - Dec 1 1990
Event16th Annual Conference of IEEE Industrial Electronics Society - IECON'90 - Pacific Grove, CA, USA
Duration: Nov 27 1990Nov 30 1990

Publication series

NameIECON Proceedings (Industrial Electronics Conference)
Volume2

Other

Other16th Annual Conference of IEEE Industrial Electronics Society - IECON'90
CityPacific Grove, CA, USA
Period11/27/9011/30/90

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Point pattern matching by a genetic algorithm'. Together they form a unique fingerprint.

Cite this