Curve Fitting Using Coevolutionary Genetic Algorithms
Curve fitting has many applications in lots of domains. The literature is full of fitting methods which are suitable for specific kinds of problems. In this paper we introduce a more general method to cover more range of problems. Our goal is to fit some cubic Bezier curves to data points of any distribution and order. The curves should be good representatives of the points and be connected and smooth. Theses constraints and the big search space make the fitting process difficult. We use the good capabilities of the coevolutionary algorithms in large problem spaces to fit the curves to the clusters of the data. The data are clustered using hierarchical techniques before the fitting process.
KeywordsCurve fitting Bezier curves coevolutionary genetic algorithms hierarchical clustering
Unable to display preview. Download preview PDF.
- 3.Manela, M., Thornhill, N., Campbell, J.: Fitting spline functions to noisy data using a genetic algorithm. In: Proceedings of the Fifth International Conference on Genetic Algorithms, pp. 549–556. Morgan Kaufmann Publishers Inc. (1993)Google Scholar
- 4.Markus, A., Renner, G., Vancza, J.: Spline interpolation with genetic algorithms. In: Proceedings of the International Conference on Shape Modeling and Applications, pp. 47–54. IEEE (2002)Google Scholar
- 5.Yin, P.: Polygonal approximation using genetic algorithms. Pattern Recognition, 838–838 (1999)Google Scholar
- 8.Wang, H., Kearney, J., Atkinson, K.: Robust and efficient computation of the closest point on a spline curve, pp. 397–406 (2002)Google Scholar