摘要
提出的自适应遗传算法采用群体的最大适应度fitmax、最小适应度fitmin、适应度平均值fitave 这 3个变量来衡量群体适应度的集中程度 ,然后根据适应度集中程度 ,自适应地变化整个群体的交叉概率pc 和变异概率pm ,改进了M .Sriniras提出的自适应遗传算法。采取最优保存策略来保证最优个体不被大的pc和pm 破坏掉。并用无放回余数随机选择算子 (RSSR选择算子 )对基本选择算子进行了改进 ,选择误差比较小。将自适应遗传算法用于图像分割的试验结果表明 ,与基本遗传算法相比 ,由于该算法综合考虑了“快速收敛”和“全局最优”这两个要求 ,因此它不仅能得到较好的分割质量 ,而且基本保持了遗传算法的运算速度 。
The genetic algorithm (GA) is derived from the mechanics of genetic adaptation in biological systems, which can search the global space of certain applications effectively. The proposed algorithm introduces three parameters, i.e. fit max , fit min and fit ave to measure how close the individuals are, thus improving the adaptive genetic algorithm (AGA) proposed by M. Sriniras. Furthermore, the elitist strategy is employed to protect the best individual of each generation, and the remainder stochastic sampling with replacement (RSSR) is employed in the proposed IAGA to improve the basic reproduction operator. The proposed IAGA is applied to image segmentation. The experimental results exhibit a satisfactory segmentation and demonstrate its learning capabilities. By determining p\-c and p\-m of the whole generation adaptively, it strikes a balance between two incompatible goals: 'sustain the global convergence capability' and 'converge rapidly to global optimum'.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2002年第5期75-78,共4页
Systems Engineering and Electronics
关键词
自适应遗传算法
交叉概率
变异概率
图像分割
启发式算法
Adaptive genetic algorithm
Crossover probability
Mutation probability
Image segmentation
