Metrology & Measurement Technology, Volume. 45, Issue 3, 70(2025)
Deep learning⁃based demodulation of Fabry⁃
Figures & Tables(12)
Fig. 4. The relationship between the wavelength of the resonant wave trough and pressure changes
Table 1.
Experimental configuration View tableView in ArticleTable 1.
Experimental configuration CPU 显卡 操作系统 开发平台 开发框架 可视化平台 Intel® Xeon® w9-3495X 1.90 GHz NVIDIA GeForce RTX 4090 Win10 Pycharm Pytorch Tensorboard
Table 2.
The results of 10⁃fold cross⁃validation for the pressure demodulation model View tableView in ArticleTable 2.
The results of 10⁃fold cross⁃validation for the pressure demodulation model 折数 训练损失 训练准确率/% 验证损失 验证准确率/% 1 1.558 1 99.68 1.554 3 99.90 2 1.547 4 99.98 1.546 8 100.00 3 1.562 1 99.24 1.563 9 98.80 4 1.640 2 90.69 1.640 0 90.80 5 1.694 2 89.90 1.689 1 90.30 6 1.636 2 90.86 1.636 5 90.90 7 1.549 2 99.99 1.548 2 100.00 8 1.632 7 90.91 1.632 7 90.90 9 1.813 2 72.71 1.812 2 72.70 10 1.634 4 90.89 1.633 6 91.00 平均 1.626 8 92.49 1.625 7 92.53
Table 3.
Comparison of experimental results of pressure demodulation models View tableView in ArticleTable 3.
Comparison of experimental results of pressure demodulation models 模型 训练 损失 训练 准确率 / % 验证 损失 验证 准确率 / % CNN 1.626 8 92.49 1.625 7 92.53 CNN⁃LSTM 1.682 8 96.98 1.679 7 96.97
Tools
Get Citation
Copy Citation Text
Hui WANG, Qichao ZHAO, Haoqi WANG, Zhiqiang SHAO, Shuang XIAO, Bin LIU. Deep learning⁃based demodulation of Fabry⁃
Paper Information
Category: Sensor Technology
Received: Feb. 17, 2025
Accepted: --
Published Online: Jul. 31, 2025
The Author Email:
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20