计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (7): 54-61.doi: 10.11896/jsjkx.250400109

• 计算机图形学 & 多媒体 • 上一篇    下一篇

单次融合随机压缩测量下的双视图分离重建方法

胡涛, 陈赞, 冯远静   

  1. 浙江工业大学信息工程学院 杭州 310014
  • 收稿日期:2025-04-23 修回日期:2025-07-19 出版日期:2026-07-15 发布日期:2026-07-10
  • 通讯作者: 陈赞(zanchen2@zjut.edu.cn)
  • 作者简介:(hutaozjut@gmail.con)
  • 基金资助:
    国家自然科学基金(U22A2040,U23A20334);浙江省“高层次人才特殊支持计划”科技创新领军人才项目(2021R52004);浙江省“尖兵”“领雁”研发攻关计划(2024C03093)

Dual-view Separation and Reconstruction Method from Fused Random Compressed Measurement

HU Tao, CHEN Zan, FENG Yuanjing   

  1. College of Information Engineering,Zhejiang University of Technology,Hangzhou 310014,China
  • Received:2025-04-23 Revised:2025-07-19 Published:2026-07-15 Online:2026-07-10
  • About author:HU Tao,born in 1999,postgraduate.His main research interests include image compressed sensing and deep learning.
    CHEN Zan,born in 1989,Ph.D,postgraduate supervisor.His main research interests include compressed sensing,sparse coding and image robust coding transmission.
  • Supported by:
    National Natural Science Foundation of China(U22A2040,U23A20334), Zhejiang Province “High-level Talent Special Support Program” Scientific and Technological Innovation(2021R52004) and Zhejiang Province Leading Geese Plan(2024C03093).

PDF (PC)

22

摘要： 压缩感知技术在图像采集与重建领域带来了革命性进展,然而针对多视图压缩感知的研究仍处于初步探索阶段,目前尚未构建出适用于单传感器、单次测量下多视图压缩重建的统一优化模型。对此,构建了一种面向双视图场景的压缩感知框架,从单次融合随机测量中有效地分离和重建两个不同场景的视图。该方法将任务分解为两个子优化问题,并引入基于近端梯度下降的迭代即插即用算法,融合图像估计与跨视图信息交互机制,通过动量反馈和残差调整实现动态信息融合。实验结果表明,与其他先进的单视角压缩感知算法相比,所提方法在低采样率下提供了更高的重建质量:在经典基准测试集Set11上、压缩率为10%的情况下,其PSNR指标高达32.19 dB。

关键词: 压缩感知, 双视图CS框架, 近端梯度下降, 信息融合

Abstract: Compressed sensing(CS) technology has brought revolutionary advancements in image acquisition and reconstruction.However,research on multi-view CS is still in its early stages,and a unified optimization model for multi-view compressed reconstruction under single-sensor,single-measurement conditions has not yet been established.This paper proposes a CS framework tailored for dual-view scenarios,which effectively separates and reconstructs two distinct scene views from a single fused random measurement.The reconstruction task is decomposed into two sub-optimization problems and addressed using an iterative plug-and-play algorithm based on proximal gradient descent,incorporating image estimation and cross-view information interaction mechanisms.Dynamic message fusion is achieved through momentum feedback and residual adjustment.Experimental results demonstrate that,compared with other advanced single-view compressed sensing algorithms,the proposed method achieves higher reconstruction quality at low sampling rates,with a maximum PSNR improvement of 2.66 dB at a 10% compression ratio on the classic benchmark dataset Set11.

Key words: Compressed sensing, Dual-view CS framework, Proximal gradient descent, Message fusion

中图分类号: 

  • TP391
[1]ZHANG Z,ZHENG S,QIU M,et al.A decade review of video compressive sensing:Aroadmap to practical applications[J].Engineering,2025,46:172-185.
[2]ZHOU Y,XU C,DAI Y,et al.Dual-view stereovision-guided automatic inspection system for overhead transmission line corridor[J].Remote Sensing,2022,14(16):4095.
[3]KIVIHARJU P.Dual View Capsule Endoscope Optics with Metallic Mirrors that Can Serve as Loop Antennas for Wireless Power Transfer[D].Sähkötekniikan korkeakoulu/ELEC,2023.
[4]ZHANG S,HUANG H,FU Y.Fast parallel implementation of dual-camera compressive hyperspectral imaging system[J].IEEE Transactions on Circuits and Systems for Video Technology,2018,29(11):3404-3414.
[5]LI X D,DUNKIN F,DEZERT J.Multi-source information fusion:Progress and future[J].Chinese Journal of Aeronautics,2024,37(7):24-58.
[6]PARK J Y,WAKIN M B.A geometric approach to multi-view compressive imaging[J].EURASIP Journal on Advances in Signal Processing,2012(1):1-15.
[7]BARANIUK R G,CEVHER V,DUARTE M F,et al.Model-based compressive sensing[J].IEEE Transactions on Information Theory,2010,56(4):1982-2001.
[8]WANG Z B,TIAN Y L,WANG R F,et al.Image Compressed Sensing Attention Neural Network Based on Residual Feature Aggregation[J].Computer Science,2023,50(4):117-124.
[9]GAN H,GAO Y,LIU C,et al.Auto BCS:Block-based imagecompressive sensing with data-driven acquisition and noniterative reconstruction[J].IEEE Transactions on Cybernetics,2021,53(4):2558-2571.
[10]DU X L,ZHU J Y,GAO X,et al.A video compressed sensing method with integrated Deformable 3D Convolution and Transformer[J].Computer Science,2025,52(11):150-156.
[11]KULKARNI K,LOHIT S,TURAGA P,et al.Reconnet:Non-iterative reconstruction of images from compressively sensed measurements[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:449-458.
[12]KIM Y,SOH J W,PARK G Y,et al.Transfer learning from synthetic to real-noise denoising with adaptive instance normalization[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:3482-3492.
[13]GUO Z,GAN H.CPP-Net:Embracing Multi-Scale Feature Fusion into Deep Unfolding CP-PPA Network for Compressive Sensing[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2024:25086-25095.
[14]KONG S,WANG W,FENG X,et al.Deep red unfolding network for image restoration[J].IEEE Transactions on Image Processing,2021,31:852-867.
[15]ZHANG Z,LIU Y,LIU J,et al.AMP-Net:Denoising-baseddeep unfolding for compressive image sensing[J].IEEE Transactions on Image Processing,2020,30:1487-1500.
[16]YUAN T H,GAN Z L.Infraredand Visible Deep Unfolding Image Fusion Network Basedon Joint Enhancement Image Pair[J].Computer Science,2024,51(10):311-319.
[17]TIAN J P,HOU B J.Compressive sensing image reconstruction based on deep unfolding self-attention network[J].Journal of Jilin University(Engineering and Technology Edition),2024,54(10):3018-3026.
[18]HOSNY S,EL-KHARASHI M W,ABDEL-HAMID A T.Survey on compressed sensing over the past two decades[J].Memories-Materials,Devices,Circuits and Systems,2023,4:100060.
[19]ZHANG Z,DENG C,LIU Y,et al.Ten-mega-pixel snapshotcompressive imaging with a hybrid coded aperture[J].Photonics Research,2021,9(11):2277-2287.
[20]QIAO M,LIU X,YUAN X.Snapshot spatial-temporal compressive imaging[J].Optics letters,2020,45(7):1659-1662.
[21]CANDES E J.The restricted isometry property and its implications for compressed sensing[J].Comptes Rendus.Mathematique,2008,346(9/10):589-592.
[22]BOOMINATHAN V,ROBINSON J T,WALLER L,et al.Recent advances in lensless imaging[J].Optica,2021,9(1):1-16.
[23]CHEN Z,HOU X,SHAO L,et al.Compressive sensing multi-layer residual coefficients for image coding[J].IEEE Transactions on Circuits and Systems for Video Technology,2019,30(4):1109-1120.
[24]ZHANG K,ZUO W,GU S,et al.Learning deep CNN denoiser prior for image restoration[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:3929-3938.
[25]LI Q,ZHOU Y,LIANG Y,et al.Convergence analysis of proximal gradient with momentum for nonconvex optimization[C]//International Conference on Machine Learning.PMLR,2017:2111-2119.
[26]CHEN B,SONG J,XIE J F,et al.Deep physics-guided unrolling generalization for compressed sensing[J].International Journal of Computer Vision,2023,131(11):2864-2887.
[27]METZLER C,MOUSAVI A,BARANIUK R.Learned D-AMP:Principled neural network based compressive image recovery[C]//NIPS.2017.
[28]CHEN Z,GUO W L,FENG Y J,et al.Deep-learned regularization and proximal operator for image compressive sensing[J].IEEE Transactions on Image Processing,2021,30:7112-7126.
[29]BOOMINATHAN V,ROBINSON J T,WALLER L,et al.Recent advances in lensless imaging[J].Optica,2021,9(1):1-16.
[30]MALEKI A,MONTANARI A.Analysis of approximate message passing algorithm[C]//2010 44th Annual Conference on Information Sciences and Systems(CISS).IEEE,2010:1-7.
[31]DONOHO D L,MALEKI A,MONTANARI A.Message-passing algorithms for compressed sensing[J].Proceedings of the National Academy of Sciences,2009,106(45):18914-18919.
[32]ZHANG K,LI Y,ZUO W,et al.Plug-and-play image restoration with deep denoiser prior[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2021,44(10):6360-6376.
[33]GAVASKAR R G,CHAUDHURY K N.Plug-and-play ISTA converges with kernel denoisers[J].IEEE Signal Processing Letters,2020,27:610-614.
[34]RAMANI S,BU T,UNSER M.Monte-carlo sure:A black-box optimization of regularization parameters for general denoising algorithms[J].IEEE Transactions on Image Processing,2008,17(9):1540-1554.
[35]METZLER C A,MALEKI A,BARANIUK R G.From denoising to compressed sensing[J].IEEE Transactions on Information Theory,2016,62(9):5117-5144.
[36]KULKARNI K,LOHIT S,TURAGA P,et al.Reconnet:Non-iterative reconstruction of images from compressively sensed measurements[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:449-458.
[37]HUANG J B,SINGH A,AHUJA N.Single image super-resolution from transformed self-exemplars[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2015:5197-5206.
[38]RAMANI S,BLU T,UNSER M.Monte-Carlo SURE:A black-box optimization of regularization parameters for general denoising algorithms[J].IEEE Transactions on Image Processing,2008,17(9):1540-1554.
[39]SHEN M,GAN H,NING C,et al.TransCS:A transformer-based hybrid architecture for image compressed sensing[J].IEEE Transactions on Image Processing,2022,31:6991-7005.
[40]WANG X,ZHAO L,ZHANG J,et al.A wavelet-domain consistency-constrained compressive sensing framework based on memory-boosted guidance filtering[J].IEEE Transactions on Instrumentation and Measurement,2024,73:1-16.
[41]XU L,MEI H,TONG F,et al.DCCT-Net:A network combined dynamic CNN and transformer for image compressive sensing[C]//IEEE International Conference on Acoustics,Speech and Signal Processing.2025:1-5.
[42]ZHANG J,ZHAO C,GAO W.Optimization-inspired compactdeep compressive sensing[J].IEEE Journal of Selected Topics in Signal Processing,2020,14(4):765-774.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发