Multi-point optical focusing based on enhanced multi-objective optimized wavefront shaping

Abstract

The phenomenon of multiple scattering has limited the development and application of many optical techniques, especially in the biomedical field. Feedback wavefront shaping methods used to overcome multiple scattering have some problems such as long optimization time. We have developed a new enhanced multi-objective optimization algorithm for wavefront shaping multi-point optical focusing, the neural network enhanced non-dominated ranking genetic algorithm (NN-NSGA2). This hybrid algorithm combines the neural network algorithm and the non-dominated ranking genetic algorithm by using the neural network predicted mask as the initial mask of the non-dominated ranking genetic algorithm, and the non-dominated ranking genetic algorithm is optimized on the basis of this initial mask. The multi-point uniform focusing of light through the scattering medium is achieved by numerical simulation and experimental study, which improves the enhancement factor value by about 10% and saves about 30% of the optimization time relative to the unenhanced multi-objective optimization algorithm, demonstrating the powerful focusing capability of the algorithm in multipoint optical focusing. Our research will have broad application prospects in many fields such as biomedicine and light manipulation in the future.