Abstract: Aiming at the problem that Harris hawks optimization algorithm converges slowly and is prone to local optimization, this paper proposes an improved Harris hawks optimization algorithm（IHHO） which combines mutually beneficial symbiosis and lens imaging learning. Firstly, the algorithm uses Tent chaotic map to initialize the population to increase the diversity of the population and improve the optimization performance of the algorithm. Secondly, in the exploration stage, the algorithm integrates the idea of mutually beneficial symbiosis and nonlinear inertia factor to enhance the exchange of population information and accelerate the convergence speed. Then, the algorithm uses lens imaging reverse learning strategy to perturb and mutate the Harris hawks position with a certain probability to improve the ability of the algorithm to jump out of the local optimum. Finally, the simulation results of 16 benchmark test functions show that IHHO has faster convergence speed, higher precision and stronger robustness compared with the other five algorithms. At the same time, IHHO is applied to the problem of image segmentation, and the simulation results verify the feasibility of the algorithm in practical engineering applications.

Key words: Harris hawks optimization, Tent chaotic map, mutually beneficial symbiosis, lens imaging, opposition-based learning, image segmentation

摘要： 针对哈里斯鹰优化算法收敛速度慢、易陷入局部最优的问题，提出一种融合互利共生和透镜成像学习的哈里斯鹰优化算法（improved Harris hawks optimization，IHHO）。利用Tent混沌映射初始化种群，增加种群多样性，提高算法寻优性能；在探索阶段融入一种互利共生思想，并引入非线性惯性因子，以增强种群信息交流，加快算法收敛速度；提出一种透镜成像反向学习策略，对哈里斯鹰位置以一定概率进行扰动变异，提高算法跳出局部最优的能力。通过16个基准测试函数进行仿真实验，结果表明，IHHO与其余5种算法相比，收敛速度更快，寻优精度更高；鲁棒性更强。同时，将IHHO应用于图像分割问题中，仿真实验验证了该算法在实际工程应用中的可行性。

关键词: 哈里斯鹰优化算法, Tent混沌映射, 互利共生, 透镜成像, 反向学习, 图像分割