图像压缩感知的特征域优化及自注意力增强神经网络重构算法
Feature-Space Optimization-Inspired and Self-Attention Enhanced Neural Network Reconstruction Algorithm for Image Compressive Sensing
查看参考文献26篇
|
文摘
|
现有的图像压缩感知(Image Compressive Sensing,ICS)优化启发网络沿用了传统算法的像素域优化思想,构建了像素域的图像信息流动通道,而没有充分利用卷积神经网络所提取的图像特征中的信息.对此,本文提出了在特征域构建信息流的思想,并设计了一种特征域优化启发ICS网络(Feature-Space Optimization-Inspired Network, FSOINet)以实现该思想.考虑到卷积操作感受野较小,本文通过将自注意力模块引入FSOINet以更高效地利用图像非局部自相似性,进一步提高重构质量,我们将其命名为FSOINet+.此外,本文还提出把迁移学习策略应用于不同采样率图像压缩感知重构网络训练中,提高网络学习效率与重构质量.仿真实验表明,本文所提出的网络在峰值信噪比(Peak Signal to Noise Ratio,PSNR)、结构相似性(Structural Similarity Index Measure,SSIM)与视觉效果上都优于现有的最优ICS重构方法,FSOINet与FSOINet+在Set11数据集上与OPINENet+相比重构图像PSNR分别平均提升了1.04 dB和1.27 dB. |
|
其他语种文摘
|
The existing optimization-inspired networks for image compressive sensing(ICS)implement information optimization and flow in the pixel domain following the traditional algorithms, which does not make full use of the information in the image feature maps extracted by the convolutional neural network. This paper proposes the idea of constructing information flow in the feature domain. A feature-space optimization-inspired network(FSOINet)is designed to implement this idea. Considering the small receptive field of the convolution operation, this paper introduces the self-attention module into FSOINet to efficiently utilize the non-local self-similarity of images to further improve the reconstruction quality, which is named FSOINet+. In addition, this paper proposes a training strategy that applies transfer learning to the ICS reconstruction network training for different sampling rates to improve the network learning efficiency and reconstruction quality. Experimental results show that the proposed method is superior to the existing state-of-the-art ICS methods in peak signal to noise ratio(PSNR), structural similarity index measure(SSIM)and the visual effect. Compared with OPINENet+ on the Set11 dataset, FSOINet and FSOINet+ have an average PSNR improvement of 1.04 dB/1.27 dB respectively. |
|
来源
|
电子学报
,2022,50(11):2629-2637 【核心库】
|
|
DOI
|
10.12263/DZXB.20220155
|
|
关键词
|
图像压缩感知
;
深度学习
;
卷积神经网络
;
自注意力
;
图像重构
;
迁移学习
|
|
地址
|
华南理工大学电子与信息学院, 广东, 广州, 510640
|
|
语种
|
中文 |
|
文献类型
|
研究性论文 |
|
ISSN
|
0372-2112 |
|
学科
|
电子技术、通信技术 |
|
基金
|
广东省自然科学基金
|
|
文献收藏号
|
CSCD:7362385
|
参考文献 共
26
共2页
|
|
1.
Donoho D L. Compressed sensing.
IEEE Transactions on Information Theory,2006,52(4):1289-1306
|
CSCD被引
3003
次
|
|
|
|
|
2.
Duarte M F. Single-pixel imaging via compressive sampling.
IEEE Signal Processing Magazine,2008,25(2):83-91
|
CSCD被引
306
次
|
|
|
|
|
3.
Zhu L. Compressed-sensing accelerated 3-dimensional magnetic resonance cholangiopancreatography: application in suspected pancreatic diseases.
Investigative Radiology,2018,53(3):150-157
|
CSCD被引
11
次
|
|
|
|
|
4.
Yuan X. Snapshot compressive imaging: theory, algorithms, and applications.
IEEE Signal Processing Magazine,2021,38(2):65-88
|
CSCD被引
21
次
|
|
|
|
|
5.
禤韵怡. 基于帧间组稀疏的两阶段递归增强视频压缩感知重构网络.
电子学报,2021,49(3):435-442
|
CSCD被引
4
次
|
|
|
|
|
6.
Gan L. Block compressed sensing of natural images.
15th International Conference on Digital Signal Processing,2007:403-406
|
CSCD被引
10
次
|
|
|
|
|
7.
Chen C. Compressedsensing recovery of images and video using multihypothesis predictions.
2011 Conference Record of the Forty Fifth Asilomar Conference on Signals, Systems and Computers,2011:1193-1198
|
CSCD被引
2
次
|
|
|
|
|
8.
Zhang J. Group-based sparse representation for image restoration.
IEEE Transactions on Image Processing,2014,23(8):3336-3351
|
CSCD被引
90
次
|
|
|
|
|
9.
Kulkarni K. Reconnet: non-iterative reconstruction of images from compressively sensed measurements.
2016 IEEE Conference on Computer Vision and Pattern Recognition,2016:449-458
|
CSCD被引
5
次
|
|
|
|
|
10.
Shi W. Deep networks for compressed image sensing.
2017 IEEE International Conference on Multimedia and Expo,2017:877-882
|
CSCD被引
2
次
|
|
|
|
|
11.
Shi W. Scalable convolutional neural network for image compressed sensing.
2019 IEEE Conference on Computer Vision and Pattern Recognition,2019:12290-12299
|
CSCD被引
1
次
|
|
|
|
|
12.
Shi W. Image compressed sensing using convolutional neural network.
IEEE Transactions on Image Processing,2020,29:375-388
|
CSCD被引
22
次
|
|
|
|
|
13.
Chen J. Learning memory augmented cascading network for compressed sensing of images.
2020 European Conference on Computer Vision,2020:513-529
|
CSCD被引
1
次
|
|
|
|
|
14.
Zhou S. Multi-channel deep networks for block-based image compressive sensing.
IEEE Transactions on Multimedia,2021,23:2627-2640
|
CSCD被引
7
次
|
|
|
|
|
15.
Zhang J. ISTA-Net: interpretable optimization-inspired deep network for image compressive sensing.
2018 IEEE Conference on Computer Vision and Pattern Recognition,2018:1828-1837
|
CSCD被引
1
次
|
|
|
|
|
16.
裴翰奇. 基于SPL迭代思想的图像压缩感知重构神经网络.
电子学报,2021,49(6):1195-1203
|
CSCD被引
3
次
|
|
|
|
|
17.
Zhang Z. AMP-Net: denoisingbased deep unfolding for compressive image sensing.
IEEE Transactions on Image Processing,2021,30:1487-1500
|
CSCD被引
15
次
|
|
|
|
|
18.
Zhang J. Optimization-inspired compact deep compressive sensing.
IEEE Journal of Selected Topics in Signal Processing,2020,14(4):765-774
|
CSCD被引
9
次
|
|
|
|
|
19.
Song J. Memory-augmented deep unfolding network for compressive sensing.
2021 ACM International Conference on Multimedia,2021:4249-4258
|
CSCD被引
1
次
|
|
|
|
|
20.
D'ascoli S. Convit: improving vision transformers with soft convolutional inductive biases.
2021 International Conference on Machine Learning,2021:2286-2296
|
CSCD被引
1
次
|
|
|
|
|
了解气象卫星更多信息，请访问