A Study of Line-Fitting Method by Using Genetic Algorithm
Takatoshi Yamagishi and Takehiko Tomikawa
Kanagawa Institute of Technology, Atsugi, Kanagawa, 243-02 Japan
A straight line approximation method of either closed curve or open-ended curve is proposed. In this system, Genetic Algorithm is considered to meet robustness requirements in a multi-point searching procedure, since conventional line fitting methods have a tendency to take troublesome parameters while having less flexibility in system configurations. Whether or not the entire divided point, which is initially allocated at regular intervals along the curved line, is adequate for approximation according to the objective function. This function: (the total length of approximate lines) – ( the value in proportion to the number of approximate points), is newly introduced in between two patterns as a quantitative measure. Some comparative simulations are performed for confirmation of this system performance. As a result, we have learned that a) the provided parameters in this system are easily handled regardless of the shape of original patterns, and b) the approximate points are not remarkably affected by the starting location of implementation along the curved line.
This article is published under a Creative Commons Attribution-NoDerivatives 4.0 Internationa License.