Adaptive Memory-Augmented Unfolding Network for Compressed Sensing
Deep unfolding networks (DUNs) have attracted growing attention in compressed sensing (CS) due to their good interpretability and high performance. However, many DUNs often improve the reconstruction effect at the price of a large number of parameters and have the problem of feature information loss...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/24/8085 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846102812092530688 |
|---|---|
| author | Mingkun Feng Dongcan Ning Shengying Yang |
| author_facet | Mingkun Feng Dongcan Ning Shengying Yang |
| author_sort | Mingkun Feng |
| collection | DOAJ |
| description | Deep unfolding networks (DUNs) have attracted growing attention in compressed sensing (CS) due to their good interpretability and high performance. However, many DUNs often improve the reconstruction effect at the price of a large number of parameters and have the problem of feature information loss during iteration. This paper proposes a novel adaptive memory-augmented unfolding network for compressed sensing (AMAUN-CS). Concretely, without loss of interpretability, we integrate an adaptive content-aware strategy into the gradient descent step of the proximal gradient descent (PGD) algorithm, driving it to adaptively capture the adequate features. In addition, we extended AMAUN-CS based on the memory storage mechanism of the human brain and propose AMAUN-CS<sup>+</sup> to develop the dependency of deep information across cascading stages. The experimental results show that the AMAUN-CS model surpasses other advanced methods on various public benchmark datasets while having lower complexity in training. |
| format | Article |
| id | doaj-art-c57a70e1771a47beaf033eda752935c6 |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-c57a70e1771a47beaf033eda752935c62024-12-27T14:52:56ZengMDPI AGSensors1424-82202024-12-012424808510.3390/s24248085Adaptive Memory-Augmented Unfolding Network for Compressed SensingMingkun Feng0Dongcan Ning1Shengying Yang2School of Information and Electronic Engineering, Zhejiang University of Science and Technology, Hangzhou 310023, ChinaSchool of Information and Electronic Engineering, Zhejiang University of Science and Technology, Hangzhou 310023, ChinaSchool of Information and Electronic Engineering, Zhejiang University of Science and Technology, Hangzhou 310023, ChinaDeep unfolding networks (DUNs) have attracted growing attention in compressed sensing (CS) due to their good interpretability and high performance. However, many DUNs often improve the reconstruction effect at the price of a large number of parameters and have the problem of feature information loss during iteration. This paper proposes a novel adaptive memory-augmented unfolding network for compressed sensing (AMAUN-CS). Concretely, without loss of interpretability, we integrate an adaptive content-aware strategy into the gradient descent step of the proximal gradient descent (PGD) algorithm, driving it to adaptively capture the adequate features. In addition, we extended AMAUN-CS based on the memory storage mechanism of the human brain and propose AMAUN-CS<sup>+</sup> to develop the dependency of deep information across cascading stages. The experimental results show that the AMAUN-CS model surpasses other advanced methods on various public benchmark datasets while having lower complexity in training.https://www.mdpi.com/1424-8220/24/24/8085compressed sensingproximal gradient descentdeep unrollingneural networksimage reconstruction |
| spellingShingle | Mingkun Feng Dongcan Ning Shengying Yang Adaptive Memory-Augmented Unfolding Network for Compressed Sensing Sensors compressed sensing proximal gradient descent deep unrolling neural networks image reconstruction |
| title | Adaptive Memory-Augmented Unfolding Network for Compressed Sensing |
| title_full | Adaptive Memory-Augmented Unfolding Network for Compressed Sensing |
| title_fullStr | Adaptive Memory-Augmented Unfolding Network for Compressed Sensing |
| title_full_unstemmed | Adaptive Memory-Augmented Unfolding Network for Compressed Sensing |
| title_short | Adaptive Memory-Augmented Unfolding Network for Compressed Sensing |
| title_sort | adaptive memory augmented unfolding network for compressed sensing |
| topic | compressed sensing proximal gradient descent deep unrolling neural networks image reconstruction |
| url | https://www.mdpi.com/1424-8220/24/24/8085 |
| work_keys_str_mv | AT mingkunfeng adaptivememoryaugmentedunfoldingnetworkforcompressedsensing AT dongcanning adaptivememoryaugmentedunfoldingnetworkforcompressedsensing AT shengyingyang adaptivememoryaugmentedunfoldingnetworkforcompressedsensing |