樊红, 刘开军, 张祖勋. 基于遗传算法的点状要素注记的整体最优配置[J]. 武汉大学学报 ( 信息科学版), 2002, 27(6): 560-565.
引用本文: 樊红, 刘开军, 张祖勋. 基于遗传算法的点状要素注记的整体最优配置[J]. 武汉大学学报 ( 信息科学版), 2002, 27(6): 560-565.
FAN Hong, LIU Kaijun, ZHANG Zuxun. A Robust Genentic Algorithm for Automated Map Name Placement[J]. Geomatics and Information Science of Wuhan University, 2002, 27(6): 560-565.
Citation: FAN Hong, LIU Kaijun, ZHANG Zuxun. A Robust Genentic Algorithm for Automated Map Name Placement[J]. Geomatics and Information Science of Wuhan University, 2002, 27(6): 560-565.

基于遗传算法的点状要素注记的整体最优配置

A Robust Genentic Algorithm for Automated Map Name Placement

  • 摘要: 提出了一种点状要素自动注记的整体最优解的解决方案,其核心算法采用具有全局搜索特性的遗传算法,可以获取点状要素注记配置的(近似)全局最优解。

     

    Abstract: The traditional algorithm for automated map name placement and its disadvantage are firstly analyzed in this paper.A new global optimization algorithms that called genetic algorithm is put forward to solve the point-labeling problem.According to the properties of map labeling,the basic design schemes and strategies of applying genetic algorithm to solve the map name placement is detailed in this paper.First,an integer vector coding scheme is adopted in the algorithm,which uses an array(or string) of integers to represent a configuration of map labeling.The initial strings can be obtained randomly by generating a random integer in a specified scope for every element(gene) of all strings.Then a fitness function of map labeling is represented as a quality evaluation function of map labeling Lastly some typical experiments are elaborated and some results obtained by the automated map-labeling program based on genetic algorithm are presented.In the meantime,a comprehensive experiment is conducted to compare this algorithm with climbing algorithm,annealing algorithm and Hopfield neural network method,and experimental results have shown that the performance of genetic algorithm is superior to those of other several traditional algorithms.

     

/

返回文章
返回