基于深度学习的Hermite-Gaussian光束模式分析方法研究  

Research on Hermite-Gaussian beam modal analysis method based on deep learning

作　　者：张杰辉 白振旭 安嘉硕 王雨雷 吕志伟 ZHANG Jiehui;BAI Zhenxu;AN Jiashuo;WANG Yuei;LU Zhiwei(School of Electronics and Information Engineering,Hebei University of Technology,Tianjin 300401,China;Hebei Key Laboratory of Advanced Laser Technology and Equipment,Tianjin 300401,China)

机构地区：[1]河北工业大学电子信息工程学院,天津300401 [2]河北省先进激光技术与装备重点实验室,天津300401

出　　处：《河北工业大学学报》2024年第5期13-21,共9页Journal of Hebei University of Technology

基　　金：国家自然科学基金资助项目(61927815,62375076);河北省自然科学基金资助项目(F2023202063);天津市自然科学基金资助项目(22JCYBJC01100);量子光学与光量子器件国家重点实验室开放课题资助项目(KF202201)。

摘　　要：提出了一种基于深度学习的Hermite-Gaussian(HG)光束模式分解方案,利用仿真生成的具有多个模式的模拟样本训练卷积神经网络,并通过训练好的模式识别网络采用迁移学习技术加速模式分解网络的训练过程。通过单次发射的强度图像,训练后的神经网络能够对四模以内所有模式的HG光束实现高效准确的模式分解,获取HG光束的模式振幅和相位信息。实验结果表明,该方法在不同模式数量的识别和分解任务中均表现出较高的精度和稳定性,有望在光通信、光学成像、激光加工等领域得到广泛应用。A deep learning-based scheme is demonstrated for the mode decomposition of Hermite-Gaussian(HG)beams.A convolutional neural network(CNN)is trained with simulated samples of multiple modes.To expedite the training process,transfer learning is employed.The trained neural network can efficiently and accurately decompose HG beams across four modes using the intensity image of a single emission,providing both mode amplitude and phase information.Experimental results demonstrate that this method achieves high accuracy and stability in recognizing and decomposing various modes.This approach holds significant potential for applications in optical communications,optical imaging,laser processing,and other fields.

关 键 词：深度学习 Hermite-Gaussian光束 模式分解 迁移学习 模式识别 

分 类 号：O439[机械工程—光学工程]