Document Type

Article

Date

1997

Embargo Period

12-19-2012

Keywords

Shape matching, Pattern recognition, Simulation annealing, Genetic algorithms, Attributed strings, Evolutionary computing

Language

English

Disciplines

Computer Sciences

Description/Abstract

Partial shape matching may be viewed as an optimization problem, to be solved using methods such as simulated annealing (SA) and genetic algorithms (GAs). We apply and compare both these methods for matching input shapes with model shapes described in terms of features such as line segments and angles. The quality of matching is gauged using a measure derived from attributed shape grammars [10, 11]. Current results show that both SA and GA succeed in the shape matching task; the GA is faster and yields the global optimum more often than the versions of SA implemented.

Share

COinS