Title :
Exploring the performance of genetic algorithms as polygonal approximators
Author :
Traver, V. Javier ; Recatala, Gabriel ; Inesta, Jose Manuel
Author_Institution :
Dept. de Inf., Univ. Jaume I, Castellon, Spain
Abstract :
The construction of polygonal approximations for digital curves is a well-known technique to obtain a compact representation of them. Over the last years, a number of methods have been proposed to optimize this process, based on different criteria. In this paper, polygonal approximation is viewed as an optimization problem, and the use of a genetic algorithm is proposed as a method to find a solution that best meets a given set of requirements. This paper analyses the capability and the performance of the genetic algorithm to select the vertices of the polygons, and some advantages and drawbacks are discussed
Keywords :
approximation theory; edge detection; genetic algorithms; image representation; digital contours; digital curves; genetic algorithms; optimization; polygonal approximation; polygons; Algorithm design and analysis; Encoding; Genetic algorithms; Iterative algorithms; Merging; Optimization methods; Performance analysis;
Conference_Titel :
Pattern Recognition, 2000. Proceedings. 15th International Conference on
Conference_Location :
Barcelona
Print_ISBN :
0-7695-0750-6
DOI :
10.1109/ICPR.2000.903657
