A symmetric forward-inverse reinforcement framework for image reconstruction through scattering media
Image retrieval from visually random optical speckles is a desired yet challenging task in various scenarios. Deep learning (DL) based approaches have rapidly grown to achieve impressive performance in recent years. However, the majority of solutions thus far have been confined to a single network t...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/179080 |
| _version_ | 1826122944175144960 |
|---|---|
| author | Qi, Pengfei Zhang, Zhengyuan Feng, Xue Lai, Puxiang Zheng, Yuanjin |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Qi, Pengfei Zhang, Zhengyuan Feng, Xue Lai, Puxiang Zheng, Yuanjin |
| author_sort | Qi, Pengfei |
| collection | NTU |
| description | Image retrieval from visually random optical speckles is a desired yet challenging task in various scenarios. Deep learning (DL) based approaches have rapidly grown to achieve impressive performance in recent years. However, the majority of solutions thus far have been confined to a single network to model the inverse scattering process, resulting in relatively poor recovery performance. In this paper, we introduce a novel objective function to embed implicit cyclic adversarial loss and propose a symmetric forward-inverse reinforcement framework congruent with this objective function for enhancing image recovery performance through scattering media, where two networks are designed to model inverse and forward scattering processes, respectively. A symmetric training strategy, aligned with the formulated objective function, is utilized to fully exploit the feature extraction ability and fine-tune the parameters of the networks, achieving higher-fidelity image recovery than that by a single neural network. Both simulation and experimental results on various datasets demonstrate the effectiveness and superiority of the proposed framework. Moreover, this framework also shows convincing robustness to varying noise levels, dataset volumes, and network parameters, indicating proficient restoration of targets from noisy speckles and maintaining comparable performance even with limited training data and fewer network parameters. Furthermore, the promising recovery results on real-world hazy datasets demonstrate that the proposed framework could open up new opportunities to enhance image restoration and recognition performance in biomedical imaging, optical encryption, and holographic display. |
| first_indexed | 2024-10-01T05:56:32Z |
| format | Journal Article |
| id | ntu-10356/179080 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T05:56:32Z |
| publishDate | 2024 |
| record_format | dspace |
| spelling | ntu-10356/1790802024-07-17T04:04:03Z A symmetric forward-inverse reinforcement framework for image reconstruction through scattering media Qi, Pengfei Zhang, Zhengyuan Feng, Xue Lai, Puxiang Zheng, Yuanjin School of Electrical and Electronic Engineering Engineering Image reconstruction Deep learning Image retrieval from visually random optical speckles is a desired yet challenging task in various scenarios. Deep learning (DL) based approaches have rapidly grown to achieve impressive performance in recent years. However, the majority of solutions thus far have been confined to a single network to model the inverse scattering process, resulting in relatively poor recovery performance. In this paper, we introduce a novel objective function to embed implicit cyclic adversarial loss and propose a symmetric forward-inverse reinforcement framework congruent with this objective function for enhancing image recovery performance through scattering media, where two networks are designed to model inverse and forward scattering processes, respectively. A symmetric training strategy, aligned with the formulated objective function, is utilized to fully exploit the feature extraction ability and fine-tune the parameters of the networks, achieving higher-fidelity image recovery than that by a single neural network. Both simulation and experimental results on various datasets demonstrate the effectiveness and superiority of the proposed framework. Moreover, this framework also shows convincing robustness to varying noise levels, dataset volumes, and network parameters, indicating proficient restoration of targets from noisy speckles and maintaining comparable performance even with limited training data and fewer network parameters. Furthermore, the promising recovery results on real-world hazy datasets demonstrate that the proposed framework could open up new opportunities to enhance image restoration and recognition performance in biomedical imaging, optical encryption, and holographic display. Ministry of Education (MOE) Nanyang Technological University This work was supported by The Ministry of Education, Singapore, under its MOE ARF Tier 2 (MOE-T2EP30123-0019). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not reflect the views of the Ministry of Education, Singapore. This work was also partially supported by National Natural Science Foundation of China (NSFC) (81930048), Hong Kong Research Grant Council (15217721), Shenzhen Science and Technology Innovation Commission (JCYJ20220818100202005). Pengfei Qi acknowledges the research scholarship awarded by the Institute of Flexible Electronics Technology of Tsinghua, Zhejiang (IFET-THU), Nanyang Technological University (NTU), and Qiantang Science and Technology Innovation Center, China (QSTIC). 2024-07-17T04:04:03Z 2024-07-17T04:04:03Z 2024 Journal Article Qi, P., Zhang, Z., Feng, X., Lai, P. & Zheng, Y. (2024). A symmetric forward-inverse reinforcement framework for image reconstruction through scattering media. Optics and Laser Technology, 179, 111222-. https://dx.doi.org/10.1016/j.optlastec.2024.111222 0030-3992 https://hdl.handle.net/10356/179080 10.1016/j.optlastec.2024.111222 2-s2.0-85195380357 179 111222 en MOE-T2EP30123-0019 Optics and Laser Technology © 2024 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies. |
| spellingShingle | Engineering Image reconstruction Deep learning Qi, Pengfei Zhang, Zhengyuan Feng, Xue Lai, Puxiang Zheng, Yuanjin A symmetric forward-inverse reinforcement framework for image reconstruction through scattering media |
| title | A symmetric forward-inverse reinforcement framework for image reconstruction through scattering media |
| title_full | A symmetric forward-inverse reinforcement framework for image reconstruction through scattering media |
| title_fullStr | A symmetric forward-inverse reinforcement framework for image reconstruction through scattering media |
| title_full_unstemmed | A symmetric forward-inverse reinforcement framework for image reconstruction through scattering media |
| title_short | A symmetric forward-inverse reinforcement framework for image reconstruction through scattering media |
| title_sort | symmetric forward inverse reinforcement framework for image reconstruction through scattering media |
| topic | Engineering Image reconstruction Deep learning |
| url | https://hdl.handle.net/10356/179080 |
| work_keys_str_mv | AT qipengfei asymmetricforwardinversereinforcementframeworkforimagereconstructionthroughscatteringmedia AT zhangzhengyuan asymmetricforwardinversereinforcementframeworkforimagereconstructionthroughscatteringmedia AT fengxue asymmetricforwardinversereinforcementframeworkforimagereconstructionthroughscatteringmedia AT laipuxiang asymmetricforwardinversereinforcementframeworkforimagereconstructionthroughscatteringmedia AT zhengyuanjin asymmetricforwardinversereinforcementframeworkforimagereconstructionthroughscatteringmedia AT qipengfei symmetricforwardinversereinforcementframeworkforimagereconstructionthroughscatteringmedia AT zhangzhengyuan symmetricforwardinversereinforcementframeworkforimagereconstructionthroughscatteringmedia AT fengxue symmetricforwardinversereinforcementframeworkforimagereconstructionthroughscatteringmedia AT laipuxiang symmetricforwardinversereinforcementframeworkforimagereconstructionthroughscatteringmedia AT zhengyuanjin symmetricforwardinversereinforcementframeworkforimagereconstructionthroughscatteringmedia |