基于图像分割的伪并行免疫遗传算法聚类设计

Study of pseudo-parallel immune genetic algorithm based on image segmentation

遗传算法是模拟生物在自然环境中的遗传和进化过程而形成的一种全局优化概率搜索算法.但该算法有时存在着早熟现象,导致搜索过早收敛,无法得到全局最优解.为此,提出了一种伪并行免疫遗传算法,在微机上利用求解问题特征以及并行思想对遗传算法的种群进行免疫接种,并进行伪并行运算,以提高搜索速度,克服早熟现象.实验结果表明,该算法具有收敛速度快,搜索精度高,稳健性强的特点.

Genetic algorithm （GA） is an algorithm used to find approximate solutions to difficult-to-solve problems through the application of the principles of evolutionary biology, to computer science. And it has the ability of doing a global searching quickly. But premature phenomenon still exists while we apply it. As a result, the searching converges earlier and we cant get the solution of the optimization. Therefore, a pseudo-parallel immune genetic algorithm is proposed, which inoculates populations generated by GA to improve the searching speed and reject premature phenomenon according to the characteristic of a certain problem and the parallel theory. The simulated results show that the solution of the optimization can be easily and quickly obtained by this method.