【摘要】 传统的图像配准的相似性测度函数对噪声过于敏感,且需要先验知识约束。对此加以改进,提出一种新的相似性测度模型。为了对模型求解,引入一种新的优化算法——细菌趋药性算法,并对其做出改进,得到多态细菌趋药性算法。实验表明,修正的相似性测度模型对噪声免疫;同时多态细菌趋药性算法比精英遗传算法、蚁群算法、粒子群算法、细菌群体趋药性算法等收敛更快,且能以更大概率收敛到全局最优。更多还原

【Abstract】 Traditional resemblance measurement function is application-restricted since it is too sensitive to noise and is subject to prior knowledge.Thus,a novel model is proposed.In order to solve the model a new optimization algorithm—Bacterial Chemotaxis Optimization(BCO) is introduced and improved as polymorphic BCO.Experiments show that the proposed model is immune to noises,and the polymorphic BCO not only converges more quickly but also has a higher probability to find global extrama than elite genetic algorithm,ant colony algorithm,particle swarm optimization,and bacterial colony chemotaxis.更多还原