Image Compressed Sensing Using Non-Local Neural Network

被引:45
|
作者
Cui, Wenxue
Liu, Shaohui
Jiang, Feng
Zhao, Debin [1 ]
机构
[1] Harbin Inst Technol, Dept Comp Sci & Technol, Harbin 150001, Peoples R China
基金
中国国家自然科学基金;
关键词
Image reconstruction; Neural networks; Reconstruction algorithms; Training; Noise reduction; Iterative algorithms; Current measurement; Convolutional neural networks (CNNs); image compressed sensing; non-local neural network; non-local self-similarity prior; THRESHOLDING ALGORITHM; RECONSTRUCTION; REGULARIZATION; FRAMEWORK;
D O I
10.1109/TMM.2021.3132489
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Deep network-based image Compressed Sensing (CS) has attracted much attention in recent years. However, the existing deep network-based CS schemes either reconstruct the target image in a block-by-block manner that leads to serious block artifacts or train the deep network as a black box that brings about limited insights of image prior knowledge. In this paper, a novel image CS framework using non-local neural network (NL-CSNet) is proposed, which utilizes the non-local self-similarity priors with deep network to improve the reconstruction quality. In the proposed NL-CSNet, two non-local subnetworks are constructed for utilizing the non-local self-similarity priors in the measurement domain and the multi-scale feature domain respectively. Specifically, in the subnetwork of measurement domain, the long-distance dependencies between the measurements of different image blocks are established for better initial reconstruction. Analogically, in the subnetwork of multi-scale feature domain, the affinities between the dense feature representations are explored in the multi-scale space for deep reconstruction. Furthermore, a novel loss function is developed to enhance the coupling between the non-local representations, which also enables an end-to-end training of NL-CSNet. Extensive experiments manifest that NL-CSNet outperforms existing state-of-the-art CS methods, while maintaining fast computational speed.
引用
收藏
页码:816 / 830
页数:15
相关论文
共 50 条
  • [21] An anisotropic non-local attention network for image segmentation
    Feiniu Yuan
    Yaowen Zhu
    Kang Li
    Zhijun Fang
    Jinting Shi
    Machine Vision and Applications, 2022, 33
  • [22] An anisotropic non-local attention network for image segmentation
    Yuan, Feiniu
    Zhu, Yaowen
    Li, Kang
    Fang, Zhijun
    Shi, Jinting
    MACHINE VISION AND APPLICATIONS, 2022, 33 (02)
  • [23] Denoised Non-Local Neural Network for Semantic Segmentation
    Song, Qi
    Li, Jie
    Guo, Hao
    Huang, Rui
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2024, 35 (05) : 7162 - 7174
  • [24] NHNet: A non-local hierarchical network for image denoising
    Zhang, Jiahong
    Cao, Lihong
    Wang, Tian
    Fu, Wenlong
    Shen, Weiheng
    IET IMAGE PROCESSING, 2022, 16 (09) : 2446 - 2456
  • [25] Video Compressed Sensing Using a Convolutional Neural Network
    Shi, Wuzhen
    Liu, Shaohui
    Jiang, Feng
    Zhao, Debin
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2021, 31 (02) : 425 - 438
  • [26] Polarimetric image denoising via non-local based cube matching convolutional neural network
    Liu, Hedong
    Li, Xiaobo
    Cheng, Zhenzhou
    Liu, Tiegen
    Zhai, Jingsheng
    Hu, Haofeng
    OPTICS AND LASERS IN ENGINEERING, 2025, 184
  • [27] Non-local Color Image Denoising with Convolutional Neural Networks
    Lefkimmiatis, Stamatios
    30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, : 5882 - 5891
  • [28] A Survey on Image Processing in Noisy Environment by Fuzzy Logic, Image Fusion, Neural Network, and Non-Local Means
    Tien, Kun-yu
    Samani, Hooman
    Hua, Jui Lui
    2017 INTERNATIONAL AUTOMATIC CONTROL CONFERENCE (CACS), 2017,
  • [29] HYPERSPECTRAL IMAGE CLASSIFICATION BASED ON NON-LOCAL NEURAL NETWORKS
    Wang, Chen
    Bai, Xiao
    Zhou, Lei
    Zhou, Jun
    2019 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2019), 2019, : 584 - 587
  • [30] Non-Local Color Compensation Network for Intrinsic Image Decomposition
    Zhang, Feng
    Jiang, Xiaoyue
    Xia, Zhaoqiang
    Gabbouj, Moncef
    Peng, Jinye
    Feng, Xiaoyi
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2023, 33 (01) : 132 - 145