- 必威体育下载

姓名
邮箱
手机号码
标题
留言内容
验证码

摘要

在微纳光纤拉锥制备过程中, 对直径的大范围、高精度和动态实时性测量是实现低损耗传输和色散调控的关键. 针对现有传统制备方法直径调控范围小、操作复杂及耗时长等问题, 本文基于深度学习神经网络算法实现了微纳光纤自动检测系统. 利用计算机视觉中的图像分割方法, 通过制作高质量多尺度微纳光纤数据集, 使用基于小目标检测改进的YOLOv8−FD算法对微纳光纤直径进行自动检测. 在数据集中获得了平均均值精度高达mAP ^IoU=50= 0.975和mAP ^IoU=50—95= 0.765的性能参数. 实验结果表明, 该系统可实现微纳光纤直径462 nm—125 μm范围, 误差2.95%以内的测量和自动化制备, 并随着光纤直径增长, 误差逐渐缩小, 且该系统光学成像单个像素分辨率为65.97 nm, 平均检测时间为9.6 ms. 本文工作适用于对微纳光纤的高精度实时测量和自动精确制备, 为低损耗传输和色散可调的微纳光纤器件发展提供新的思路.

关键词:

Abstract

The wide range, high precision, and dynamic real-time measurement of micro-/nanofiber diameter are crucial for achieving low loss transmission and controlling dispersion in the preparation process of micro-nanofiber. In view of the problems of small diameter regulation range, complex operation and long-time consumption of the existing preparation methods, the automatic detection system of micro-nanofiber is realized based on deep learning neural network algorithm in this work. The image segmentation method in computer vision is used to make high-quality multi-scale micro-/nanofiber datasets, and the improved YOLOv8-FD (You Only Look Once version 8-Fiber Detection) algorithm based on small target detection is used to automatically detect the diameter of micro-nanofiber. Through image segmentation and identification of the target of single pixel size in the microscopic image, the diameter detection of micro-nanofiber is finally realized. In this process, the real-time diameter of micro-nanofiber is obtained through image information, and then the micro-nanofiber small target is accurately segmented to achieve the precise detection of mAP ^IoU ⁼⁵⁰= 0.975 and mAP ^IoU ^=50—95= 0.765 on the micro-nanofiber multi-scale target dataset with extremely high accuracy. The algorithm-based construction of a high-precision micro-nanofiber automatic preparation system enables real-time accurate segmentation of fiber edges, calculation of fiber diameter, and feedback to the control system for achieving automated preparation of fibers with arbitrary diameters. Additionally, it facilitates the detection of micro-nanofiber in a range from 462 nm to 125 μm. The average response time for reasoning is 9.6 ms, and the detection error is kept below 2.95%. In addition, compared with other micro-/nanofiber diameter detection methods based on optical imaging and mode cutoff, this method shows advantages of high precision, high speed and arbitrary diameter preparation for diameter detection based on deep learning neural networks. The system is very suitable for high-precision real-time measurement and automatic and accurate preparation of micro-/nanofibers, thereby providing a novel method of developing micro-nanofiber devices with low-loss transmission and adjustable dispersion.

Keywords:

作者及机构信息

通信作者:谷付星,gufuxing@usst.edu.cn

基金项目:国家自然科学基金(批准号: 62122054, 62075131)和上海市教委“曙光计划”(批准号: 22SG44)资助的课题.

Authors and contacts

Corresponding author:Gu Fu-Xing,gufuxing@usst.edu.cn

Funds:Project supported by the National Natural Science Foundation of China (Grant Nos. 62122054, 62075131) and the “Dawn” Program of Shanghai Education Commission, China (Grant No. 22SG44).

文章全文: translate this paragraph

参考文献

[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]

施引文献

下载: 全尺寸图片幻灯片

图 2 　(a)微纳光纤自动制备系统示意图; (b)微纳光纤自动制备流程图

下载: 全尺寸图片幻灯片

标定次数	1	2	3	4
像素个数/pixel	845.01	843.52	844.25	843.24
标定系数/(nm·pixel^–1)	65.89	66.01	65.95	66.03

下载: 导出CSV

光纤图示	$ {{d}}_{1}/ $μm	AFM图示	$ {{d}}_{2}/ $μm	$ {\varepsilon }/{\%} $
图5(d)	0.489	图5(a)	0.475	2.95
图5(e)	0.731	图5(b)	0.714	2.38
图5(f)	1.935	图5(c)	1.907	1.47

下载: 导出CSV

[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]

[1]	吴婉玲, 王向珂, 虞华康, 李志远.基于微纳光纤双模式干涉的亚波长聚焦光场及光捕获应用. 必威体育下载 , 2024, 73(10): 100401.doi:10.7498/aps.73.20240181
[2]	欧秀娟, 肖奕.RNA扭转角预测的深度学习方法. 必威体育下载 , 2023, 72(24): 248703.doi:10.7498/aps.72.20231069
[3]	孙涛, 袁健美.基于深度学习原子特征表示方法的Janus过渡金属硫化物带隙预测. 必威体育下载 , 2023, 72(2): 028901.doi:10.7498/aps.72.20221374
[4]	战庆亮, 葛耀君, 白春锦.基于深度学习的流场时程特征提取模型. 必威体育下载 , 2022, 71(7): 074701.doi:10.7498/aps.71.20211373
[5]	张航瑛, 王雪琦, 王华英, 曹良才.基于明度分量的Retinex-Net图像增强改进方法. 必威体育下载 , 2022, 71(11): 110701.doi:10.7498/aps.71.20220099
[6]	徐昭, 周昕, 白星, 李聪, 陈洁, 倪洋.基于深度学习的相位截断傅里叶变换非对称加密系统攻击方法. 必威体育下载 , 2021, 70(14): 144202.doi:10.7498/aps.70.20202075
[7]	王甜甜, 王慧, 朱艳春, 王丽嘉.基于位移流U-Net和变分自动编码器的心脏电影磁共振图像左心肌运动追踪. 必威体育下载 , 2021, 70(22): 228701.doi:10.7498/aps.70.20210885
[8]	南虎, 麻晓晶, 赵海博, 汤少杰, 刘卫华, 王大威, 贾春林.基于YOLOv3框架的高分辨电镜图像原子峰位置检测. 必威体育下载 , 2021, 70(7): 076803.doi:10.7498/aps.70.20201502
[9]	苏博, 陶芬, 李可, 杜国浩, 张玲, 李中亮, 邓彪, 谢红兰, 肖体乔.同步辐射纳米CT图像配准方法研究. 必威体育下载 , 2021, 70(16): 160704.doi:10.7498/aps.70.20210156
[10]	张瑶, 张云波, 陈立.基于深度学习的光学表面杂质检测. 必威体育下载 , 2021, 70(16): 168702.doi:10.7498/aps.70.20210403
[11]	赵智鹏, 周双, 王兴元.基于深度学习的新混沌信号及其在图像加密中的应用. 必威体育下载 , 2021, 70(23): 230502.doi:10.7498/aps.70.20210561
[12]	郎利影, 陆佳磊, 于娜娜, 席思星, 王雪光, 张雷, 焦小雪.基于深度学习的联合变换相关器光学图像加密系统去噪方法. 必威体育下载 , 2020, 69(24): 244204.doi:10.7498/aps.69.20200805
[13]	陈炜, 郭媛, 敬世伟.基于深度学习压缩感知与复合混沌系统的通用图像加密算法. 必威体育下载 , 2020, 69(24): 240502.doi:10.7498/aps.69.20201019
[14]	李杰, 李蒙蒙, 孙立朋, 范鹏程, 冉洋, 金龙, 关柏鸥.保偏微纳光纤倏逝场传感器. 必威体育下载 , 2017, 66(7): 074209.doi:10.7498/aps.66.074209
[15]	吴涛, 金义富, 侯睿, 杨俊杰.不确定性边缘表示与提取的认知物理学方法. 必威体育下载 , 2013, 62(6): 064201.doi:10.7498/aps.62.064201
[16]	刘颖刚, 车伏龙, 贾振安, 傅海威, 王宏亮, 邵敏.微纳光纤布拉格光栅折射率传感特性研究. 必威体育下载 , 2013, 62(10): 104218.doi:10.7498/aps.62.104218
[17]	宋长新, 马克, 秦川, 肖鹏.结合稀疏编码和空间约束的红外图像聚类分割研究. 必威体育下载 , 2013, 62(4): 040702.doi:10.7498/aps.62.040702
[18]	侯建平, 赵晨阳, 杨楠, 郝建苹, 赵建林.微纳光纤端面反射特性的实验测量方法. 必威体育下载 , 2013, 62(14): 144216.doi:10.7498/aps.62.144216
[19]	梁瑞冰, 孙琪真, 沃江海, 刘德明.微纳尺度光纤布拉格光栅折射率传感的理论研究. 必威体育下载 , 2011, 60(10): 104221.doi:10.7498/aps.60.104221
[20]	唐英干, 邸秋艳, 赵立兴, 关新平, 刘福才.基于二维最小Tsallis交叉熵的图像阈值分割方法. 必威体育下载 , 2009, 58(1): 9-15.doi:10.7498/aps.58.9

计量

文章访问数:1520
PDF下载量:98
被引次数:0

搜索

留言板