光学精密工程2025，Vol.33Issue(15)：2331-2341,11.DOI:10.37188/OPE.20253315.2331
深度学习融合模型共轭涡旋光干涉微位移测量
Conjugate vortex beam interferometry for micro-displacement measurement with deep learning fusion architecture
摘要
Abstract
To overcome the limitations of traditional fringe-based displacement inversion algorithms in vor-tex interferometry for micro-displacement measurement,a deep-learning fusion model based on conjugate vortex beam interference is proposed.A YOLOv8s-Seg segmentation network,incorporating a light-weight FasterNet backbone and a CARAFE dynamic upsampling module,is employed to segment petal regions in interference images accurately,thereby reducing the influence of background noise and beam dis-tortion on phase information extraction.A 14-layer convolutional neural network(CNN)is then used to perform multi-scale hierarchical feature extraction on the segmented petal regions,establishing a precise mapping between morphological variations and rotation angles to enable sub-nanometer displacement de-tection.Experimental results within a displacement range of(0-500)nm demonstrate a petal-region seg-mentation mean average precision(mAP)of 96.5%,overall displacement accuracy better than 0.94 nm,and a mean absolute error(MAE)of 0.63 nm.Owing to the dual-network collaborative architecture,the proposed method exhibits improved robustness to fringe distortion and noise,providing marked advantages in both precision and stability for micro-displacement measurement.关键词
微位移测量/共轭涡旋光干涉/YOLOv8s-Seg分割网络/多尺度分层特征提取Key words
micro-displacement measurement/conjugate vortex light interferometry/YOLOv8s-Seg seg-mentation/multi-scale hierarchical feature extraction分类
数理科学引用本文复制引用
杨雪娇,刘吉,武锦辉,袁涛,王仕杰,姬翔峰,于丽霞,张博洋,陈相..深度学习融合模型共轭涡旋光干涉微位移测量[J].光学精密工程,2025,33(15):2331-2341,11.基金项目
山西省基础研究计划资助项目(No.202203021211087,No.202203021221101,No.202403021222181) （No.202203021211087,No.202203021221101,No.202403021222181）
山西省回国留学人员科研资助项目(No.2023-141,No.2024-117) （No.2023-141,No.2024-117）
山西省科技成果转化引导专项(No.202404021301029) （No.202404021301029）
