基于自学习局部线性嵌入的多幅亚像元超分辨成像

穆绍硕; 张叶; 贾平

doi:10.3788/OPE.20152309.2677

您当前的位置：

首页 >

文章列表页 >

基于自学习局部线性嵌入的多幅亚像元超分辨成像

信息科学 | 更新时间：2020-08-12

- 基于自学习局部线性嵌入的多幅亚像元超分辨成像
- Super-resolution imaging of multi-frame sub-pixel images based on self-learning LLE
- 光学精密工程 2015年23卷第9期页码：2677-2686
- 作者机构：
  
  1. 中国科学院大学北京,中国,100049
  2. 中国科学院长春光学精密机械与物理研究所中国科学院航空光学成像与测量重点实验室,吉林长春,130033
- 作者简介：
- 基金信息：
  
  吉林省重大科技攻关项目(No.11ZDGG001)();国家自然科学基金青年基金资助项目(No.60902067)()
- DOI：10.3788/OPE.20152309.2677
  中图分类号： TP73;TP391
- 收稿日期：2014-10-28，
  
  修回日期：2014-12-05，
  
  纸质出版日期：2015-09-25
- 稿件说明：
移动端阅览
穆绍硕, 张叶, 贾平. 基于自学习局部线性嵌入的多幅亚像元超分辨成像[J]. 光学精密工程, 2015,23(9): 2677-2686

MU Shao-shuo, ZHANG Ye, JIA Ping. Super-resolution imaging of multi-frame sub-pixel images based on self-learning LLE[J]. Editorial Office of Optics and Precision Engineering, 2015,23(9): 2677-2686
穆绍硕, 张叶, 贾平. 基于自学习局部线性嵌入的多幅亚像元超分辨成像[J]. 光学精密工程, 2015,23(9): 2677-2686 DOI： 10.3788/OPE.20152309.2677.

MU Shao-shuo, ZHANG Ye, JIA Ping. Super-resolution imaging of multi-frame sub-pixel images based on self-learning LLE[J]. Editorial Office of Optics and Precision Engineering, 2015,23(9): 2677-2686 DOI： 10.3788/OPE.20152309.2677.

摘要

研究了软硬件相结合的亚像元超分辨成像技术。首先通过探测器扫描获得同一场景彼此错位亚像元像素的多帧图像作为训练样本和输入图像;然后针对传统的局部线性嵌入(LLE)实例学习超分辨算法过于依赖外部训练样本

不利于光电成像系统直接处理等缺点

提出了一种基于自学习的改进LLE算法;采用新的LLE权值计算方法获得正数权值

同时对初始估计再次运用自学习LLE方法恢复丢失的高频细节信息。仿真实验结果表明

该算法重构的图像的信噪比比传统LLE超分辨算法提高了0.8 dB

运行时间提高了75%

视觉上可感知重构图像的细节信息更丰富。与其它方法相比

用搭载的微位移实验平台运行本文算法所获得重构图像的信噪比和信息熵都有很大提高

表明本文算法能获得高质量和高分辨率的重构图像。

Abstract

A super-resolution technology of combining hardware and software was researched. Firstly

the detector scanning was used to obtain multiple images with the same scene produced by different motion parameters and they were chosen to be training sets and input images. In consideration of that traditional Local Linear Embedding(LLE) super-resolution technology is over-relying on external training images and is inconvenient for processing image directly

a improved self-learning algorithm based on the LLE was proposed. The new LLE weight calculation method was proposed to obtain initial estimation of HR image. Meanwhile

self-learning LLE algorithm was used to recover lost high-frequency information of initial estimation and to obtain the final estimation. Simulation results show that the Peak Signal to Noise Radio(PSNR) of the reconstructed image by proposed algorithm improves 0.8 dB and operation time shortens by 75% as compared with those of conventional LLE method

respectively. Moreover

in the real scene experiment of micro-displacement platform

the Signal to Noise Radio (SNR) and information entropy of the reconstructed image by proposed algorithm have also greatly improved as compared with those of other algorithms. The algorithm provides high quality reconstruction image and improves the resolution of the captured image.

关键词

Keywords

references

WEN D S, LIU X P, QIAO W, et al.. Novel subpixel imaging system with linear CCD sensors[J]. Sensors and Controls for Intelligent Manufacturing Ⅱ, 2001, 4563:116-122.

刘新平,王虎,汶德胜. 亚象元线阵CCD焦平面的光学拼接[J]. 光子学报,2002,31(6):781-784. LIU X P, WANG H, WEN D SH. Optical focal plane assembly of linear CCD array for subpixel imaging camera [J]. Acta Photonica Sinica, 2002,31(6):781-784.(in Chinese)

徐正平,翟林培,葛文奇,等. 亚像元的CCD几何超分辨方法[ J]. 光学精密工程,2008, 16(12) :2447-2453. XU ZH P, ZHAI L P, GE W Q, et al.. CCD geometric super-resolution method based on subpixel [J]. Opt. Precision Eng.,2008,16(12):2447-2453.(in Chinese)

杨文波, 朱明, 刘志明, 等. 基于3线阵探测器的亚像元成像超分辨率重构[J]. 光学精密工程,2014,22(8):2247-2258. YANG W B, ZHU M, LIU Z M, et al.. Super-resolution reconstruction of sub-pixel imaging achieved by three linear array detectors [J].Opt. Precision Eng., 2014, 22(8):2247-2258.(in Chinese)

VANDEWALLE P, SUSSTRUNK S, VETTERLI M. A frequency domain approach to registration of aliased images with application to super-resolution[J]. Eurasip Journal on Applied Signal Processing, 2006,71459.

IRANI M, PELEG S. Improving resolution by image registration[J].Graphical Model and Image Processing, 1991,53(3):231-239.

CHEESEMAN P, KANEFSKY B, KRAFT R, et al.. Super-resolved surface reconstruction from multiple images[C] . 13th International Workshop on Maximum Entropy and Bayesian Methods(MAXENT 93) Santa Barbara, California, 1993.

HARDIE R C, BARNARD K J, ARNSTRONG E E. Joint MAP registration and high-resolution image estimation using a sequence of under-sampled images[J]. Proc of the IEEE Conference on Image Processing, 1997, 6(12):1621-1633.

FARSIU S, ROBINSON M D, ELAD M, et al.. Fast and robust multi-frame super resolution[J]. Proc of the IEEE Conference on Image Processing, 2004, 13(10):1327-1344.

PROTTER M, ELAD M, TAKEDA H, et al.. Generalizing the nonlocal-means to super-resolution reconstruction[J]. Proc of the IEEE Conference on Image Processing, 2009, 18(1):36-51.

李家德,张叶,贾平. 采用非局部均值的超分辨率重构[J]. 光学精密工程, 2013,21(12):1576-1585. LI J D, ZHANG Y, JIA P. Super-resolution reconstruction using nonlocal means[J]. Opt. Precision Eng., 2013,21(12):1576-1585.(in Chinese)

SHIN C J, YEONG K. Super-resolution algorithm using noise level adaptive dictionary[C]. IEEE 14th International Symposium on Consumer Electronics,2010.

RAMAKANTH S A, BABU R V. Super-resolution using a single image dictionary[C]. 2014 IEEE International Conference on Electronics, Computing and Communication Technologies(CONECCT), 2014:1-6.

YANG J, WRIGHT J, HUANG T S, et al.. Image super-resolution as sparse representation of raw image patches[C]. Proc of the IEEE Conference on Computer Vision and Pattern Recognition:CVPR, 2008:1-8.

YANG J, WRIGHT J, HUANG T S, et al.. Image super-resolution via sparse representation[J]. IEEE Trans. Image Process,ZERETZKI R, 2010,19(11):2861-2873.

HE L, QI H R, ZERETZKI R, et al. Beta process joint dictionary learning for coupled feature spaces with application to single image super-resolution[C]. Proc of the IEEE Conference on Computer Vision and Pattern Recognition:CVPR, 2013.

ROWEIS S, SAUL L. Nonlinear dimensionality reduction by locally linear embedding[J]. Science, 2000,290(5500):2323-2326.

彭真明, 景亮, 何艳敏, 等. 基于多尺度稀疏字典的多聚焦图像超分辨融合[J]. 光学精密工程, 2014, 22(1):169-176. PENG ZH M, JING L, HE Y M, et al.. Superresolution fusion of multi-focus image based on multiscale sparse dictionary [J]. Opt. Precision Eng., 2014, 22(1):169-176.(in Chinese)

CHANG H, YEUNG D Y, XIONG Y. Super-resolution through neighbor embedding[C]. CVPR, 2004,1:275-282.

TANIGUCHI K, OHASHI M,HAN XH, et al.. Example-based super-resolution using locally linear embedding[C]. Proceedings of the Sixth International Conference on Computer Sciences and Convergence Information Technology:(ICCIT), 2011:861-865.

LIU Q J, LIU L N, WANG Y H, et al.. Locally linear embedding based example learning for pan-sharpening[C]. 21st International Conference on Pattern Recognition (ICPR 2012), Tsukuba, Japan, 2012:1928-1931.

KAIBING Z, XINBO G, XUELONG L, et al.. Partially supervised neighbor embedding for example-based image super-resolution[J]. IEEE Journal of Selected Topics in Signal Processing, 2011,5(2):230-239.

TAO Y, ZONGLIANG G, XIUCHANG Z. Novel neighbor embedding super resolution method for compressed images[C]. Proceedings of the 2012 International Conference on Image Analysis and Signal Processing: (IASP), 2012:1-4.

BEVILACQUA M, ROUMY A, GUILEMOT C, et al.. Neighbor embedding based single-image super-resolution using semi-nonnegative matrix factorization[J]. IEEE International Conference on Acoustics, Speech and Signal Processing, 2012,1:1289-1292.

KAIBING Z, XINBO G, DACHENG T, et al..Single image super-resolution with multiscale similarity learning [J]. IEEE Transactions on Neural Networks and Learning Systems, 2013,10(24):1648-1659.

SU K, TIAN Q, XUE Q, et al.. Neighborhood issue in single-frame image super-resolution[C]. 2005 IEEE International Conference on Multimedia and Expo(ICME), The Netherlands, 2005:1122-1125.

HENG L. Variational local structure estimation for image super-resolution[C].In proceeding of the IEEE International Conference on Image Processing, Atlanta, GA, Oct. 2006:1721-1724.

浏览量

266

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

面向超分辨光学成像的浸没微球透镜控制

利用异形像元探测器的超分辨成像方法

一种具有自学习功能的高速目标识别方法