摘要
根据个体适应度大小和变化快慢来自适应改变遗传算法中的计算参数,并引入基于直方图二阶导数的约束条件,对遗传算法进行了两方面的优化,不仅可增加计算的自适应程度以增大计算结果的准确性,而且可缩小搜索空间,提高运算速度,从而形成一种基于优化遗传算法的模糊C均值聚类图像分割算法。实验结果证明,文中算法不仅减小了最小均方误差,改善了分割效果,而且大大缩短了计算时间,提高了运算效率。
The genetic algorithm was improved in the following aspects ： （1） the parameters in the genetic algorithm were adjusted adaptively according to the value and the varying velocity of individual fitness to increase the genetic algorithm＇s adaptability and the accuracy of results; （2） the constraint based on the second order derivative of histogram was introduced into geneticalgorithm to reduce the searching scope and increase calculating efficiency. Consequently, a novel fuzzy clustering image segmentation algorithm based on improved genetic algorithm was proposed. The experimental results show that the segmentation algorithm, compared with the method using： simple genetic algorithm, costs less time, reduces mean square error and improves segmentation results.
出处
《弹箭与制导学报》
CSCD
北大核心
2008年第4期190-192,198,共4页
Journal of Projectiles,Rockets,Missiles and Guidance
基金
国家高等学校博士学科点专项科研基金(20040699015)
西北工业大学研究生创业种子基金(Z200632)资助
关键词
图像分割
遗传算法
模糊聚类
image segmentation
genetic algorithm
fuzzy clustering
