提出一种通用式突变算子用于增强反馈式波前整形系统的调控效率，进而实现激光透过散射介质后的高效聚焦。为验证该突变算子提高聚焦效率的有效性，在经典优化算法，包括遗传算法、粒子种群算法、蚁群算法、模拟退火算法等四种算法的基础上引入突变算子，以优化结束后的增强因子和达到最高增强因子时的迭代周期数来表征聚焦效率。经过数值仿真和实验验证，该突变算子的引入使得四种经典优化算法的聚焦效率均得到大幅提升，增强因子提升了25%以上，同时迭代周期数减少了63%以上。当增加调控单元数量时，突变算子的高效性将更为显著。为进一步验证该突变算子的通用性，对二元振幅型调制以及多点聚焦进行了数值模拟分析，结果表明该突变算子有效增强了聚焦效率。该研究为反馈式波前整形的多种经典算法与多种调控方式提供了更高效的聚焦策略，实现了散射介质后更快更强的光斑聚焦，在光捕获、光遗传学等领域具有潜在的应用价值。

When light passes through scattering media, such as biological tissues and multimode fibers, the wavefront of the beam is disturbed due to multiple scattering and distortion. This phenomenon is usually seen as an obstacle to biomedical imaging, telecommunications, and photodynamic therapy. As an effective method, iterative wavefront shaping is capable of manipulating the incident wavefront and compensating the wavefront distortion due to multiple scattering. Recent advances in iterative wavefront shaping techniques have made it possible to manipulate the light focusing and transport in scattering media. To improve the optimization performance, various optimization algorithms and improved strategies have been utilized. However, an improved strategy that is suitable for various algorithms has not been demonstrated yet. Here, a novel guided mutation strategy is proposed to improve optimization efficiency for light focusing through scattering medium. Not limited to a specific algorithm, guided mutation strategy is extended to various feedback-based wavefront shaping algorithms. In this study, single point focusing is firstly conducted in a feedback wavefront shaping system based on multiple classical optimization algorithms, including genetic algorithm, particle swarm algorithm, ant colony algorithm and simulated annealing algorithm. To validate the effectiveness of the guided mutation strategy in improving the focusing efficiency, the guided mutation strategy is introduced on the basis of the above four algorithms. The focusing efficiency is characterized by the enhancement factor after optimization and the number of iteration cycles when the maximum enhancement factor is reached as made by regular algorithms. Through numerical simulation and experimental verification, the guided mutation strategy greatly improves the focusing efficiency of the four classical optimization algorithms. The enhancement factor increases by more than 25%, the number of iteration cycles is reduced by more than 63%. When the input modes numbers increases, the benefits of the guided mutation strategy will become more significant. To further verify the universality of the guided mutation strategy, numerical simulation analysis of single point focusing with binary modulation and multi-point focusing with multi-objective genetic algorithm are also carried out. The results show that, similar to the single point focusing with phase modulation, the guided mutation strategy can effectively enhance the focusing efficiency with binary modulation and multi-objective optimization. This investigation of binary and multi-objective optimization further demonstrate that guided mutation strategy can be applied to widely applications, such as binary amplitude optimization system and multi-point uniform focusing. Overall, this study provides a more efficient focusing strategy for various classical algorithms and regulation methods of feedback wavefront shaping. For both phase modulation and binary amplitude modulation, considerable improvements in optimization effect and rate have been obtained with the introduce of guided mutation strategy. Because of the effectiveness and universality of the guided mutation strategy, it will be beneficial for applications ranging from controlling the transmission of light through disordered media to optical manipulation behind them. And this research will have potential application value in the field of fiber laser, two-photon microscopy and optogenetics.