Focusing through disturbed multimode optical fiber based on self-adaptive genetic algorithm

In terms of the imaging resolution and miniaturization, multimode fiber outperforms single-mode fiber bundles. However, fiber disturbance has always been a challenge to multimode fiber imaging. The great majority of existing algorithms are incapable of dealing with abrupt and strong perturbations from outside. In this paper, a self-adaptive genetic algorithm is proposed. When the environment changes rapidly, additional new elements can be introduced, preventing the optimization from being trapped in a local optimum. The effectiveness of the algorithm is verified by simulation. The experimental results indicate that it enables the speckle pattern to refocus better when the fiber is perturbed. The algorithm can also be used to transmit images. Our research will facilitate the advancement of detection technology using multimode optical fiber in the industrial and medical domains.