基于量子空间的蚁群算法及应用

Ant colony algorithm and application based on quantum space

针对蚁群算法收敛速度慢,容易陷入局部极值的缺点,提出将量子进化算法与蚁群算法相融合的新算法。在该算法中,蚂蚁当前位置用量子比特的两个概率幅表示,与普通蚁群算法相比,个体数量相等时,新算法的搜索空间将加倍,同时用量子非门来实现变异操作,相比传统算法,在寻优过程中具有更好的种群多样性并有效克服了蚁群算法的早熟及停滞现象。将此算法用于图像分割,实验结果表明,该方法有效解决了蚁群算法收敛速度慢和容易陷入局部极值的问题,而且在分割速度和精度上得到了较大提高。

Aimming at the low convergent rate of the ant colony algorithm （ACA） and the disadvantage of falling into local extremum easily, this paper proposes a method of combining quantum evolutionary algorithms with ACA, which regards two probability amplitudes of quantum bits as current location of ant. When the num ber of ants is the same, the proposed algorithm makes the search space double and uses the quantum gate to re alize the variation operation. Compared with the traditional algorithms, it has better population diversity in the optimization process and effectively avoids the prematurity and stagnation phenomenon of ACA. This method can be used in image segmentation. The experimental results show that this method is effective to solve the slow convergence rate and easy to fall into local extremum problems of ACA, and segmentation speed and precision have been improved greatly.