Filters








15 Hits in 5.0 sec

DR2-Net: Deep Residual Reconstruction Network for Image Compressive Sensing [article]

Hantao Yao, Feng Dai, Dongming Zhang, Yike Ma, Shiliang Zhang, Yongdong Zhang, Qi Tian
2017 arXiv   pre-print
In this paper, we propose a novel Deep Residual Reconstruction Network (DR^2-Net) to reconstruct the image from its Compressively Sensed (CS) measurement.  ...  Most traditional algorithms for compressive sensing image reconstruction suffer from the intensive computation.  ...  RELATED WORK This work is related with image compressive sensing image reconstruction, deep learning for compressive sensing, and the deep residual network.  ... 
arXiv:1702.05743v4 fatcat:vwaa6b2o2bf6lbbje2xbbp2kxi

Cascaded Reconstruction Network for Compressive image sensing [article]

Yahan Wang, Huihui Bai, Lijun Zhao, Yao Zhao
2018 arXiv   pre-print
In CSRNet, deep reconstruction network module obtains an initial image with acceptable quality, which can be further improved by residual network module based on convolutional neural network.  ...  The first reconstruction network is a compatibly sampling reconstruction network (CSRNet), which recovers an image from its compressively sensed measurement sampled by a traditional random matrix.  ...  Both the sampling module and its matching residual reconstruction module form a complete compressive sensing image reconstruction network, named ASRNet.  ... 
arXiv:1712.03627v2 fatcat:lfvelk5qr5bmhmt27hdurmcymu

ATP-Net: An Attention-based Ternary Projection Network For Compressed Sensing [article]

Guanxiong Nie, Yajian Zhou
2021 arXiv   pre-print
Furthermore, a compressed sensing algorithm especially for image reconstruction is implemented, on the basis of the ternary sampling matrix, which is called ATP-Net, i.e., Attention-based ternary projection  ...  Compressed Sensing (CS) theory simultaneously realizes the signal sampling and compression process, and can use fewer observations to achieve accurate signal recovery, providing a solution for better and  ...  ATTENTION-BASED TERNARY PROJECTION FOR COMPRESSED SENSING In this section, the proposed end-to-end deep learning framework for compressed sensing, namely Attention-based ternary projection network (ATP-Net  ... 
arXiv:2106.12728v1 fatcat:y7533tb2tzdw5bsqdbdwbmwjyi

A Review of Image Compressed Sensing in Deep Learning

Kaiguo Xia, Lei Hu, Pengqiang Mao
2020 DEStech Transactions on Engineering and Technology Research  
In recent years, deep learning has developed rapidly in the field of image recognition, which provides a new idea for the reconstruction of compressed sensing.  ...  This paper sorts out the current image compressed sensing reconstruction methods based on deep learning, analyzes the characteristics and key steps of the algorithm according to three different deep network  ...  Due to the addition of residual layers in Dr2-Net, the computational complexity was greatly increased.  ... 
doi:10.12783/dtetr/mcaee2020/35015 fatcat:76cw4vfu7fcafmxtfeqv73kwmq

Compressed Sensing Image Reconstruction Based on Convolutional Neural Network

Yuhong Liu, Shuying Liu, Cuiran Li, Danfeng Yang
2019 International Journal of Computational Intelligence Systems  
In order to solve these problems and further improve the quality of image processing, a new convolutional neural network structure CombNet is proposed, which uses the measured value of compression sensing  ...  A B S T R A C T Compressed sensing theory is widely used in image and video signal processing because of its low coding complexity, resource saving, and strong anti-interference ability.  ...  However, the performance of the traditional compression sensing reconstruction algorithm is not perfect.  ... 
doi:10.2991/ijcis.d.190808.001 fatcat:amqyrlxvnjf7ppuvlshsc3ljgu

A two-branch convolution residual network for image compressive sensing

Chenquan Gan, Xiaoqin Yan, Yunfeng Wu, Zufan Zhang
2019 IEEE Access  
INDEX TERMS Image compressive sensing, two-branch convolution, residual network, structural similarity, reconstruction performance, visual quality.  ...  For better CS reconstruction, the image is preliminarily reconstructed by the deconvolution decoder network, and then the residual network is used to optimize the pre-reconstructed image.  ...  CONCLUSION In this paper, a two-branch convolution residual network for image compressive sensing (denoted as TCR-CS) has been proposed.  ... 
doi:10.1109/access.2019.2961369 fatcat:o6536i5xe5hdrkaygkjntb3qou

Research on Image Reconstruction of Compressed Sensing Based on a Multi-Feature Residual Network

Ruili Nan, Guiling Sun, Zhihong Wang, Xiangnan Ren
2020 Sensors  
The experimental results show that the compressed sensing image reconstruction method based on the multi-feature residual network proposed in this paper can improve the quality of crop image reconstruction  ...  features of the crop image to realize deep reconstruction of the image, and complete the inverse solution of compressed sensing.  ...  DR2-Net [6] uses the residual module to solve the problem of deep network degradation and achieve image reconstruction at low measurement rate. Bora et al.  ... 
doi:10.3390/s20154202 pmid:32731604 fatcat:mu6ze6aibjfzvhkbrmaxttv3em

Full Image Recover for Block-Based Compressive Sensing [article]

Xuemei Xie, Chenye Wang, Jiang Du, Guangming Shi
2018 arXiv   pre-print
Recent years, compressive sensing (CS) has improved greatly for the application of deep learning technology. For convenience, the input image is usually measured and reconstructed block by block.  ...  This usually causes block effect in reconstructed images. In this paper, we present a novel CNN-based network to solve this problem.  ...  EXPERIMENTS In this section, we conduct the experiments on MSE loss with fully convolutional neural network for compressive sensing problem.  ... 
arXiv:1802.00179v1 fatcat:22srmtwekrfmhabmrjkmbmhqbu

MC-ISTA-Net: Adaptive Measurement and Initialization and Channel Attention Optimization inspired Neural Network for Compressive Sensing [article]

Nanyu Li, Cuiyin Liu
2019 arXiv   pre-print
The optimization inspired network can bridge convex optimization and neural networks in Compressive Sensing (CS) reconstruction of natural image, like ISTA-Net+, which mapping optimization algorithm: iterative  ...  in optimization-inspired networks.  ...  Kuldeep [12] proposed ReconNet, using convolutional network to handle CS reconstruction. Further, Hantao Yao [13] proposed DR2-Net, which uses the residual network to deal with CS.  ... 
arXiv:1902.09878v3 fatcat:3zrblmqtzzaofit7hktu4vtxtu

AMPA-Net: Optimization-Inspired Attention Neural Network for Deep Compressed Sensing [article]

Nanyu Li, Charles C. Zhou
2020 arXiv   pre-print
Compressed sensing (CS) is a challenging problem in image processing due to reconstructing an almost complete image from a limited measurement.  ...  To achieve fast and accurate CS reconstruction, we synthesize the advantages of two well-known methods (neural network and optimization algorithm) to propose a novel optimization inspired neural network  ...  Neural network: a fast implicit model for CS Reconstruction, such as DR2-Net [12] , Recon-Net [13] , Adaptive-Recon-Net [14] .  ... 
arXiv:2010.06907v6 fatcat:wl3udgecxjesffuuhzgyvvtmpq

Multi-scale Residual Reconstruction Neural Network with Non-local Constraint

Wan Li, Fang Liu, Licheng Jiao, Fei Hu
2019 IEEE Access  
INDEX TERMS Compressive sensing, residual neural network, non-local constraint, multi-scale, deep learning. 70910 2169-3536  ...  With the development of the neural network, some novel reconstructed networks are proposed to solve the problem of compressive sensing (CS) reconstruction.  ...  Then, Deep Residual Reconstruction Network (DR2-Net) is proposed in [17] to improve the quality of the reconstructed image base on the residual learning.  ... 
doi:10.1109/access.2019.2918593 fatcat:tihidavngjfitdmbaw7jgi4xya

Block-based compressed sensing of MR images using multi-rate deep learning approach

Ejaz Ul Haq, Huang Jianjun, Xu Huarong, Kang Li
2021 Complex & Intelligent Systems  
In this work, we proposed multi-rate method using deep neural networks for block-based compressive sensing of magnetic resonance images with performance that greatly outperforms existing state-of-the-art  ...  In block-based compressive sensing, a number of deep models are needed to train with each corresponding to different sampling rate.  ...  [31] proposed deep dense connected network for the compressive sensing MRI.  ... 
doi:10.1007/s40747-021-00426-6 fatcat:fe7xvi2hwfaiph26tx4qltjifa

An Entropy-Based Algorithm with Nonlocal Residual Learning for Image Compressive Sensing Recovery

Zhonghua Xie, Lingjun Liu, Cui Yang
2019 Entropy  
Image recovery from compressive sensing (CS) measurement data, especially noisy data has always been challenging due to its implicit ill-posed nature, thus, to seek a domain where a signal can exhibit  ...  In this paper, we propose a novel entropy-based algorithm for CS recovery to enhance image sparsity through learning the group sparsity of residual.  ...  Either pure convolutional layers (Deep Inverse [22] ) or a combination of convolutional and fully connected layers (DR2-Net [23] and ReconNet [24] ) is used to build deep learning frameworks.  ... 
doi:10.3390/e21090900 fatcat:ccdgs3vqbfeprdgysqvbozkiea

Compressive Sensing Radar Imaging with Convolutional Neural Networks

Qiao Cheng, Achintha Avin Ihalage, Yujie Liu, Yang Hao
2020 IEEE Access  
ACKNOWLEDGMENTS The authors would also like to thank Dr Andre Andy and Geoff Simpson for their help in setting up the MIMO array imaging system.  ...  [15] introduced a deep residual reconstruction network (DR2-Net) that consists of a linear mapping network for a high quality preliminary image first to be reconstructed and then further improved by  ...  More recently, several deep neural networks have been proposed for CS image reconstruction [13] - [15] .  ... 
doi:10.1109/access.2020.3040498 fatcat:enw26oavzrcxxj5htjss7cnq7i

Computational Methods for Image Acquisition and Analysis with Applications in Optical Coherence Tomography

Fangliang Bai
2021
This method can achieve better accuracy, hardware-efficient image acquisition and reconstruction than the conventional compressive sensing algorithm.  ...  We further proposed a convolutional neural network model, LSHR-Net, as the first deep-learning imaging solution for the single-pixel camera.  ...  Conclusions This chapter presented a mixed-weights deep neural network solution for image reconstruction from compressively sensed measurements.  ... 
doi:10.22024/unikent/01.02.89414 fatcat:ffplclehqbd6jirskxhgfvn6a4