Intelligent phase imaging guided by physics models
Implicit neural representation characterizes the mapping between the signal’s coordinate to its attributes, and has been widely used in the optimization of inverse problems by embedding the physics process into the loss function.However, the implicit neural representation is suffering the low conver...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
China InfoCom Media Group
2023-06-01
|
| Series: | 物联网学报 |
| Subjects: | |
| Online Access: | http://www.wlwxb.com.cn/zh/article/doi/10.11959/j.issn.2096-3750.2023.00345/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841533779747799040 |
|---|---|
| author | Zhen LIU Hao ZHU You ZHOU Zhan MA Xun CAO |
| author_facet | Zhen LIU Hao ZHU You ZHOU Zhan MA Xun CAO |
| author_sort | Zhen LIU |
| collection | DOAJ |
| description | Implicit neural representation characterizes the mapping between the signal’s coordinate to its attributes, and has been widely used in the optimization of inverse problems by embedding the physics process into the loss function.However, the implicit neural representation is suffering the low convergence speed and accuracy from the random initialization of the network parameters.The meta-learning algorithm for providing implicit neural representation with a strong prior of network parameters was proposed, thus enhancing the optimization efficiency and accuracy for solving inverse problems.To address the important issue of lens less phase imaging, an intelligent method on phase imaging was proposed based on the snapshot lens less sensing model.By embedding the optical diffraction propagation theory into the design of loss function for implicit neural representation, the dependency of large-scale labelled dataset in traditional deep learning-based methods was eliminated and accurate phase image from a single diffraction measurement image was provided.Furthermore, the meta-learning model was introduced for initializing network to further improve the efficiency and accuracy of network training.Numerical simulation results show that the proposed method can achieve a PSNR improvement of more than 11 dB compared to the conventional method.The experimental results in real data show that the phase image reconstructed by the proposed method is clearer and has fewer artifacts. |
| format | Article |
| id | doaj-art-0a2954af5d3e4bb59da3453a5e8332c7 |
| institution | Kabale University |
| issn | 2096-3750 |
| language | zho |
| publishDate | 2023-06-01 |
| publisher | China InfoCom Media Group |
| record_format | Article |
| series | 物联网学报 |
| spelling | doaj-art-0a2954af5d3e4bb59da3453a5e8332c72025-01-15T02:54:30ZzhoChina InfoCom Media Group物联网学报2096-37502023-06-017354259577797Intelligent phase imaging guided by physics modelsZhen LIUHao ZHUYou ZHOUZhan MAXun CAOImplicit neural representation characterizes the mapping between the signal’s coordinate to its attributes, and has been widely used in the optimization of inverse problems by embedding the physics process into the loss function.However, the implicit neural representation is suffering the low convergence speed and accuracy from the random initialization of the network parameters.The meta-learning algorithm for providing implicit neural representation with a strong prior of network parameters was proposed, thus enhancing the optimization efficiency and accuracy for solving inverse problems.To address the important issue of lens less phase imaging, an intelligent method on phase imaging was proposed based on the snapshot lens less sensing model.By embedding the optical diffraction propagation theory into the design of loss function for implicit neural representation, the dependency of large-scale labelled dataset in traditional deep learning-based methods was eliminated and accurate phase image from a single diffraction measurement image was provided.Furthermore, the meta-learning model was introduced for initializing network to further improve the efficiency and accuracy of network training.Numerical simulation results show that the proposed method can achieve a PSNR improvement of more than 11 dB compared to the conventional method.The experimental results in real data show that the phase image reconstructed by the proposed method is clearer and has fewer artifacts.http://www.wlwxb.com.cn/zh/article/doi/10.11959/j.issn.2096-3750.2023.00345/implicit neural representationphysics modelphase imagingmeta learningunsupervised learning |
| spellingShingle | Zhen LIU Hao ZHU You ZHOU Zhan MA Xun CAO Intelligent phase imaging guided by physics models 物联网学报 implicit neural representation physics model phase imaging meta learning unsupervised learning |
| title | Intelligent phase imaging guided by physics models |
| title_full | Intelligent phase imaging guided by physics models |
| title_fullStr | Intelligent phase imaging guided by physics models |
| title_full_unstemmed | Intelligent phase imaging guided by physics models |
| title_short | Intelligent phase imaging guided by physics models |
| title_sort | intelligent phase imaging guided by physics models |
| topic | implicit neural representation physics model phase imaging meta learning unsupervised learning |
| url | http://www.wlwxb.com.cn/zh/article/doi/10.11959/j.issn.2096-3750.2023.00345/ |
| work_keys_str_mv | AT zhenliu intelligentphaseimagingguidedbyphysicsmodels AT haozhu intelligentphaseimagingguidedbyphysicsmodels AT youzhou intelligentphaseimagingguidedbyphysicsmodels AT zhanma intelligentphaseimagingguidedbyphysicsmodels AT xuncao intelligentphaseimagingguidedbyphysicsmodels |