基于停滞检测的双向搜索灰狼优化算法

针对灰狼优化算法（GWO）易陷入局部最优、收敛速度低的问题，提出了一种基于停滞检测的双向搜索灰狼优化算法（DBGWO）。为了提升初始种群的质量，引入了Bernouilli shift映射；为了充分利用GWO特有的头狼机制，实现整体提升算法性能的目的，提出一种双向搜索策略；为了提升算法跳出局部最优的能力、增加算法的收敛速度，提出一种停滞检测机制，针对算法是否有陷入局部最优风险的判断，狼群会采取相应的措施改变当前状态。通过对23个基准测试函数进行仿真实验结果表明，所提算法在求解多峰函数问题上效果显著，同时在求解最优解非0点的函数问题上表现也较为优越。将该算法用于求解多阈值图像分割问题，解决了用Kapur熵法计算多阈值时耗时过长的问题。

Bidirectional search grey wolf optimizer based on stagnation detection

Aiming at the problem that grey wolf optimizer(GWO) was easy to fall into local optimization and low convergence speed, this paper proposed a bidirectional search grey wolf optimizer based on stagnation detection(DBGWO). In order to improve the quality of the initial population, it introduced Bernouilli shift mapping. In order to make full use of the unique wolf mechanism of GWO and achieve the purpose of improving the performance of the algorithm as a whole, it proposed a bi-directional search strategy. In order to improve the ability of jumping out of local optimization and increase the convergence speed of the algorithm, it proposed a stagnation detection mechanism. According to the judgment of whether the algorithm had the risk of falling into the local optimization, the wolves would take corresponding measures to change the current state. Through simulation experiments on 23 benchmark functions, the results show that this algorithm is effective in solving multi-modal function problems, and it is also superior in solving function problems with non-zero points of the optimal solution. In addition, this algorithm was used to solve the problem of multi-threshold image segmentation, which solved the problem that the Kapur entropy method was too time-consuming to calculate the multi-threshold.