Adaptive Optical Closed-Loop Control on the Basis of Hyperparametric Optimization of Convolutional Neural Networks
Abstract
:1. Introduction
2. Fundamental Principles
2.1. Principles and Methods of Wavefront Sensorless Control Systems in the Basis of Convolutional Neural Networks
2.2. Principles of Hyper-Parameter Optimization Convolutional Neural Network
3. Analysis and Discussion
3.1. Numerical Simulation
3.2. Numerical Analysis and Discussion
3.3. Experimental and Result Analysis
4. Conclusions
Author Contributions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
Sample Availability
References
- Nishizaki, Y.; Valdivia, M.; Horisaki, R.; Kitaguchi, K.; Saito, M.; Tanida, J.; Vera, E. Deep learning wavefront sensing. Opt. Express 2019, 27, 240–251. [Google Scholar] [CrossRef] [PubMed]
- Toselli, I.; Gladysz, S. Improving system performance by using adaptive optics and aperture averaging for laser communications in oceanic turbulence. Opt. Express 2020, 28, 17347–17361. [Google Scholar] [CrossRef]
- Sun, L.; Guo, Y.; Shao, C.; Li, Y.; Zheng, Y.; Sun, C.; Wang, X.; Huang, L. 10.8 kW, 2.6 times diffraction limited laser based on a continuous wave Nd: YAG oscillator and an extra-cavity adaptive optics system. Opt. Lett. 2018, 43, 4160–4163. [Google Scholar] [CrossRef] [PubMed]
- Pomohaci, R.; Oudmaijer, R.D.; Goodwin, S.P. A pilot survey of the binarity of Massive Young Stellar Objects with K-band adaptive optics. Mon. Not. R. Astron. Soc. 2019, 484, 226–238. [Google Scholar] [CrossRef] [Green Version]
- Azimipour, M.; Migacz, J.V.; Zawadzki, R.J.; Werner, J.S.; Jonnal, R.S. Functional retinal imaging using adaptive optics swept-source OCT at 1.6 MHz. Optica 2019, 6, 300–303. [Google Scholar] [CrossRef] [PubMed]
- Salter, P.S.; Booth, M.J. Adaptive optics in laser processing. Light Sci. Appl. 2019, 8, 1–16. [Google Scholar] [CrossRef] [Green Version]
- Chen, J.; Ma, J.; Mao, Y.; Liu, Y.; Li, B.; Chu, J. Experimental evaluation of a positive-voltage-driven unimorph deformable mirror for astronomical applications. Opt. Eng. 2015, 54, 117103. [Google Scholar] [CrossRef]
- Zhao, F.; Huang, W.; Xu, W.; Yang, T. Optimization method for the centroid sensing of Shack-Hartmann wavefront sensor. Infrared Laser Eng. 2014, 43, 3005–3009. [Google Scholar]
- Barchers, J.; Fried, D.; Link, D. Evaluation of the performance of Hartmann sensors in strong scintillation. Appl. Opt. 2002, 41, 1012–1021. [Google Scholar] [CrossRef]
- Wei, P.; Li, X.; Luo, X. Influence of lack of light in partial subapertures on wavefront reconstruction for shack-Hartmann wavefront sensor. Chin. J. Lasers 2020, 47, 0409002. [Google Scholar]
- Wu, Z.; Zhang, T.; Mbemba, D.; Wang, Y.; Wei, X.; Zhang, Z.; Iqbal, A. Wavefront Sensorless Aberration Correction with Magnetic Fluid Deformable Mirror for Laser Focus Control in Optical Tweezer System. IEEE Trans. Magn. 2020, 57, 4600106. [Google Scholar] [CrossRef]
- Booth, M.J. Wavefront sensorless adaptive optics for large aberrations. Opt. Lett. 2007, 32, 5–7. [Google Scholar] [CrossRef]
- Huang, L.H. Coherent beam combination using a general model-based method. Chin. Phys. Lett. 2014, 31, 094205. [Google Scholar] [CrossRef]
- Liu, M.; Dong, B. Efficient wavefront sensorless adaptive optics based on large dynamic crosstalk-free holographic modal wavefront sensing. Opt. Express 2022, 30, 9088–9102. [Google Scholar] [CrossRef]
- Yang, H.; Li, X. Comparison of several stochastic parallel optimization algorithms for adaptive optics system without a wavefront sensor. Opt. Laser Technol. 2011, 43, 630–635. [Google Scholar] [CrossRef]
- Ren, H.; Dong, B. Fast dynamic correction algorithm for model-based wavefront sensorless adaptive optics in extended objects imaging. Opt. Express 2021, 29, 27951–27960. [Google Scholar] [CrossRef]
- Paine, S.W.; Fienup, J.R. Machine learning for improved image-based wavefront sensing. Opt. Lett. 2018, 43, 1235–1238. [Google Scholar] [CrossRef] [Green Version]
- Swanson, R.; Lamb, M.; Correia, C.; Sivanandam, S.; Kutulakos, K. Wavefront reconstruction and prediction with convolutional neural networks. In Proceedings of the Adaptive Optics Systems VI; SPIE: Austin, TX, USA, 2018; Volume 10703, pp. 481–490. [Google Scholar]
- Hu, L.; Hu, S.; Gong, W.; Si, K. Deep learning assisted Shack–Hartmann wavefront sensor for direct wavefront detection. Opt. Lett. 2020, 45, 3741–3744. [Google Scholar] [CrossRef]
- Tan, M.; Le, Q. Efficientnet: Rethinking model scaling for convolutional neural networks. In Proceedings of the International Conference on Machine Learning, PMLR, Beach, CA, USA, 10–15 June 2019; pp. 6105–6114. [Google Scholar]
- Wang, M.; Guo, W.; Yuan, X. Single-shot wavefront sensing with deep neural networks for free-space optical communications. Opt. Express 2021, 29, 3465–3478. [Google Scholar] [CrossRef]
- Landman, R.; Haffert, S.Y.; Radhakrishnan, V.M.; Keller, C.U. Self-optimizing adaptive optics control with reinforcement learning for high-contrast imaging. J. Astron. Telesc. Instruments Syst. 2021, 7, 039002. [Google Scholar] [CrossRef]
- Tian, Q.; Lu, C.; Liu, B.; Zhu, L.; Pan, X.; Zhang, Q.; Yang, L.; Tian, F.; Xin, X. DNN-based aberration correction in a wavefront sensorless adaptive optics system. Opt. Express 2019, 27, 10765–10776. [Google Scholar] [CrossRef] [PubMed]
- Masum, M.; Shahriar, H.; Haddad, H.; Faruk, M.J.H.; Valero, M.; Khan, M.A.; Rahman, M.A.; Adnan, M.I.; Cuzzocrea, A.; Wu, F. Bayesian hyperparameter optimization for deep neural network-based network intrusion detection. In Proceedings of the 2021 IEEE International Conference on Big Data (Big Data), Orlando, FL, USA, 15–18 December 2021; pp. 5413–5419. [Google Scholar]
- Chalasani, S.R.; Selvam, G.T.; Chellaiah, C.; Srinivasan, A. Optimized convolutional neural network-based multigas detection using fiber optic sensor. Opt. Eng. 2021, 60, 127108. [Google Scholar]
- Feurer, M.; Hutter, F. Hyperparameter optimization. In Automated Machine Learning; Springer: Cham, Switzerland, 2019; pp. 3–33. [Google Scholar]
- Victoria, A.H.; Maragatham, G. Automatic tuning of hyperparameters using Bayesian optimization. Evol. Syst. 2021, 12, 217–223. [Google Scholar] [CrossRef]
- Yang, L.; Shami, A. On hyperparameter optimization of machine learning algorithms: Theory and practice. Neurocomputing 2020, 415, 295–316. [Google Scholar] [CrossRef]
- Yang, N.; Nan, L.; Zhang, D.; Ku, T. Research on image interpretation based on deep learning. Infrared Laser Eng. 2018, 47, 203002–0203002. [Google Scholar] [CrossRef]
- Neiswanger, W.; Wang, K.A.; Ermon, S. Bayesian algorithm execution: Estimating computable properties of black-box functions using mutual information. In Proceedings of the International Conference on Machine Learning, PMLR, Virtual Event, 18–24 July 2021; pp. 8005–8015. [Google Scholar]
- Ellinger, J.; Beck, L.; Benker, M.; Hartl, R.; Zaeh, M.F. Automation of Experimental Modal Analysis Using Bayesian Optimization. Appl. Sci. 2023, 13, 949. [Google Scholar] [CrossRef]
Hyperparameters | Value Range of the Hyperparameters (Integer, Floating Point, Boolean, and Character) |
---|---|
Number of network stacking layers | 3–12 |
Network width of each layer | 4–40 |
Convolution kernel size | 1–7 |
Convolution step size | 1–3 |
Fill or not | 0 or 1 |
Pooling mode | Max Pooling, verage Pooling, Mix Pooling, Stochastic Pooling |
Activation function type | Sigmoid, Tanh, ReLu, Leaky, ReLu, ELU |
The number of neurons in the adequately connected layer | 256–1024 |
Type of optimization algorithm | Adam, Sgd |
Optimized the learning rate of the algorithm | 0.0001–0.01 |
Types of Models | Test Set Optimal RMSE (V) | Number of Model Parameters |
---|---|---|
Auto-U-net | 0.5476 V | 5,946,094 |
Resnet18 | 3.1584 V | 58,430,240 |
Inception v3 | 1.1312 V | 21,249,511 |
U-net | 0.898 V | 40,492,546 |
EfficientNet-B0 | 1.2569 V | 7,247,472 |
DenseNet | 1.864 V | 12,687,072 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Chen, B.; Zhou, Y.; Jia, J.; Zhang, Y.; Li, Z. Adaptive Optical Closed-Loop Control on the Basis of Hyperparametric Optimization of Convolutional Neural Networks. Appl. Sci. 2023, 13, 8589. https://doi.org/10.3390/app13158589
Chen B, Zhou Y, Jia J, Zhang Y, Li Z. Adaptive Optical Closed-Loop Control on the Basis of Hyperparametric Optimization of Convolutional Neural Networks. Applied Sciences. 2023; 13(15):8589. https://doi.org/10.3390/app13158589
Chicago/Turabian StyleChen, Bo, Yilin Zhou, Jingjing Jia, Yirui Zhang, and Zhaoyi Li. 2023. "Adaptive Optical Closed-Loop Control on the Basis of Hyperparametric Optimization of Convolutional Neural Networks" Applied Sciences 13, no. 15: 8589. https://doi.org/10.3390/app13158589
APA StyleChen, B., Zhou, Y., Jia, J., Zhang, Y., & Li, Z. (2023). Adaptive Optical Closed-Loop Control on the Basis of Hyperparametric Optimization of Convolutional Neural Networks. Applied Sciences, 13(15), 8589. https://doi.org/10.3390/app13158589