Multi-Scale Channel Distillation Network for Image Compressive Sensing
Recently, convolutional neural networks (CNNs) have demonstrated striking success in computer vision tasks. Methods based on CNNs for image compressive sensing (CS) have also gained prominence. However, existing methods tend to increase the depth of the network in feature space for better reconstruc...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10835084/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832592968417017856 |
|---|---|
| author | Tianyu Zhang Kuntao Ye Yue Zhang Rui Lu |
| author_facet | Tianyu Zhang Kuntao Ye Yue Zhang Rui Lu |
| author_sort | Tianyu Zhang |
| collection | DOAJ |
| description | Recently, convolutional neural networks (CNNs) have demonstrated striking success in computer vision tasks. Methods based on CNNs for image compressive sensing (CS) have also gained prominence. However, existing methods tend to increase the depth of the network in feature space for better reconstruction quality, neglecting the hierarchical representation of intermediate features in pixel space. In order to coordinate the feature space and pixel space to complete the deep reconstruction of images, and further improve the reconstruction performance of current CS methods, we propose a multi-scale channel distillation network (MSCDN). This network first obtains images of multiple scales using a scale-space image decomposition method at the sampling stage, followed by sampling these decomposed images through a convolutional operation. In this way, multi-scale information in the compressed domain is aggregated. During the reconstruction phase, a low-frequency information recovery network generates a preliminary image, whereas a high-frequency feature aggregation network refines the image further. Specifically, we design a dual-branch deep reconstruction architecture with channel distillation residual block (CDRB) as the core component. One branch extracts features gradually by cascading multiple CDRB modules, thereby supplementing the initial reconstructed image with a large amount of high-frequency content in feature space. The other branch takes the initial reconstructed image as input and sequentially fuses the intermediate feature outputs by CDRBs to increase the local details of the image in pixel space. Combining outputs from both branches, we achieve an optimal reconstructed image. Extensive experimental results on four benchmark datasets demonstrate that MSCDN surpasses state-of-the-art CS methods not only in reconstruction accuracy but also in perceptual visual quality. |
| format | Article |
| id | doaj-art-e80d390fc07b424cac11794d691f7465 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-e80d390fc07b424cac11794d691f74652025-01-21T00:01:45ZengIEEEIEEE Access2169-35362025-01-01139524953710.1109/ACCESS.2025.352775610835084Multi-Scale Channel Distillation Network for Image Compressive SensingTianyu Zhang0https://orcid.org/0009-0008-7781-0366Kuntao Ye1https://orcid.org/0000-0001-5780-6310Yue Zhang2https://orcid.org/0009-0003-2669-8945Rui Lu3https://orcid.org/0000-0002-8998-7184School of Science, Jiangxi University of Science and Technology, Ganzhou, ChinaSchool of Science, Jiangxi University of Science and Technology, Ganzhou, ChinaSchool of Science, Jiangxi University of Science and Technology, Ganzhou, ChinaElectric Power Communication Department, State Grid Xinyu Power Supply Company, Xinyu, ChinaRecently, convolutional neural networks (CNNs) have demonstrated striking success in computer vision tasks. Methods based on CNNs for image compressive sensing (CS) have also gained prominence. However, existing methods tend to increase the depth of the network in feature space for better reconstruction quality, neglecting the hierarchical representation of intermediate features in pixel space. In order to coordinate the feature space and pixel space to complete the deep reconstruction of images, and further improve the reconstruction performance of current CS methods, we propose a multi-scale channel distillation network (MSCDN). This network first obtains images of multiple scales using a scale-space image decomposition method at the sampling stage, followed by sampling these decomposed images through a convolutional operation. In this way, multi-scale information in the compressed domain is aggregated. During the reconstruction phase, a low-frequency information recovery network generates a preliminary image, whereas a high-frequency feature aggregation network refines the image further. Specifically, we design a dual-branch deep reconstruction architecture with channel distillation residual block (CDRB) as the core component. One branch extracts features gradually by cascading multiple CDRB modules, thereby supplementing the initial reconstructed image with a large amount of high-frequency content in feature space. The other branch takes the initial reconstructed image as input and sequentially fuses the intermediate feature outputs by CDRBs to increase the local details of the image in pixel space. Combining outputs from both branches, we achieve an optimal reconstructed image. Extensive experimental results on four benchmark datasets demonstrate that MSCDN surpasses state-of-the-art CS methods not only in reconstruction accuracy but also in perceptual visual quality.https://ieeexplore.ieee.org/document/10835084/Convolutional neural networkscompressive sensingintermediate featuremulti-scaleimage reconstruction |
| spellingShingle | Tianyu Zhang Kuntao Ye Yue Zhang Rui Lu Multi-Scale Channel Distillation Network for Image Compressive Sensing IEEE Access Convolutional neural networks compressive sensing intermediate feature multi-scale image reconstruction |
| title | Multi-Scale Channel Distillation Network for Image Compressive Sensing |
| title_full | Multi-Scale Channel Distillation Network for Image Compressive Sensing |
| title_fullStr | Multi-Scale Channel Distillation Network for Image Compressive Sensing |
| title_full_unstemmed | Multi-Scale Channel Distillation Network for Image Compressive Sensing |
| title_short | Multi-Scale Channel Distillation Network for Image Compressive Sensing |
| title_sort | multi scale channel distillation network for image compressive sensing |
| topic | Convolutional neural networks compressive sensing intermediate feature multi-scale image reconstruction |
| url | https://ieeexplore.ieee.org/document/10835084/ |
| work_keys_str_mv | AT tianyuzhang multiscalechanneldistillationnetworkforimagecompressivesensing AT kuntaoye multiscalechanneldistillationnetworkforimagecompressivesensing AT yuezhang multiscalechanneldistillationnetworkforimagecompressivesensing AT ruilu multiscalechanneldistillationnetworkforimagecompressivesensing |