|本期目录/Table of Contents|

[1]李翠华*,施华,戴平阳,等.引入格式塔理论的超分辨率图像重建技术[J].厦门大学学报(自然科学版),2011,50(02):261.
 LI Cui hua*,SHI Hua,DAI Ping yang,et al.Superresolution Image Reconstruction Based on Gestalt Theory[J].Journal of Xiamen University(Natural Science),2011,50(02):261.
点击复制

引入格式塔理论的超分辨率图像重建技术(PDF)
分享到:

《厦门大学学报(自然科学版)》[ISSN:0438-0479/CN:35-1070/N]

卷:
50卷
期数:
2011年02期
页码:
261
栏目:
研究论文
出版日期:
2011-02-28

文章信息/Info

Title:
Superresolution Image Reconstruction Based on Gestalt Theory
作者:
李翠华*施华戴平阳陈婧杜晓凤曲延云谢怡
厦门大学信息科学与技术学院,福建 厦门 361005
Author(s):
LI Cuihua*SHI HuaDAI PingyangCHEN Jing DU XiaofengQU YanyunXIE Yi
School of Information Science and Technology,Xiamen University,Xiamen 361005,China
关键词:
格式塔理论超分辨率图像重建低可观测目标正则性约束
Keywords:
Gestalt theorysuperresolutionimage reconstructionlow observable targetregularization restriction
分类号:
TP 391.41
文献标志码:
-
摘要:
超分辨率图像重建是一个不适定问题,学术上富有挑战性,在影像处理、高分辨率对地观测等领域具有广泛的用途,其目标评价特性与格式塔理论关于视像认知的相以性、共性线、同趋性等高度吻合.本文引入格式塔理论,在国际图像与视频压缩标准JPEG2000、MPEG4推荐的提升小波分解与重建框架下,开展重建级大于分解级的超分辨率图像重建模型与算法研究,有如下三方面的贡献:1) 对高频信息进行高精度估计,重建获得超分辨率图像,最高频系数置零的小波变换去噪为超分辨率图像重建提供了一个逆向成功的范例;2) 遵循格式塔理论,对边缘轮廓、纹理及高频细节信息进行重建;3) 建立了带格式塔约束的超分辨率图像重建优化模型框架,体现了人类视觉评价指标.
Abstract:
Superresolution (SR) image reconstruction is an illposed which has been widely used in some challenging fields such as image processing and highresolution for earth observation.The evaluation criteria of SR image reconstruction in terms of Video recognition like vicinity,similarity,and continuity of direction,are identical with the ones of Gestalt theory.Using the decomposition and reconstruction techniques of lifting wavelet recommended by JPEG2000 and MPEG4,this paper applies Gestalt theory to study the model of SR image reconstruction in which the level of reconstruction is higher than the level of decomposition.Our main contributions include:1) Estimate the high frequency information with high precision,reconstruct SR image and provide a successful reverse example of SR image reconstruction which adopts wavelet denoising and sets the coefficient of the highest frequency as zero;2) Reconstruct the edge contour,texture and high frequency information according to Gestalt theory; 3) Set up an optimal model for SR image reconstruction with the Gestalt constraints,which reflects the evaluation criteria of human vision.

参考文献/References:


[1]Van Ouwerkerk J D.Image superresolution survey[J].Image and Vision Computing,2006,24:10391052.
[2]Palmer S E.Modern theories of Gestalt perception[J].Mind & Language,1990,5(4):289293.
[3]Harris J L.Diffraction and resolving power[J].Journal of the Optical Society of America,1964,54 (7):931936.
[4]Goodman J W.Introduction to Fourier optics[M].New York:McGrawHill,1968.
[5]Chaudhuri S.Superresolution imaging[M].Boston:Kluwer Academic Publishers,2001.
[6]Park S,Park M,Kang M.Superresolution image reconstruction:a technical overview[J].IEEE Signal Processing Magazine,2003,20(3):2136.
[7]Ng M K,Bose N K.Mathematical analysis of superresolution methodology[J].IEEE Signal Processing Magazine,2003,20(3):6274. [8]Kang M G,Chaudhuri S.Special issue on superresolution image reconstruction[J].IEEE Signal Processing Magazine,2003,20(3):1920.
[9]Tsai R Y,Huang T S.Multiframe image restoration and registration[C]// Advances in Computer Vision and Image Processing Magazine.Greewich,CT:JAI Press Inc,1984:317339.
[10]Rhee S H,Kang M G.Discrete cosine transformbased regularized highresolution image reconstruction algorithm[J].SPIE Journal Optical Engineering,1999,38(8):13481356.
[11]Nguyen N,Milanfar P.A waveletbased interpolationrestoration method for superresolution (wavelet superresolution)[J].Circuits Systems and Signal Processing,2000,19(4):321338.
[12]Chan R H,Chan T F,Shen L X,et al.Wavelet algorithms for highresolution image reconstruction[J].SIAM Journal on Scienctific Computing,2003,24(4):14081432.
[13]Bose N K,Chappalli M B.A secondgeneration wavelet framework for superresolution with noise filtering[J].International Journal of Image Systems and Technology,2004,14(2):8489.
[14]Schultz R R,Stevenson R L.Extraction of highresolution frames from video sequences[J].IEEE Trans on Image Processing,1996,5(6):9961011.
[15]Stark H,Oskoui P.Highresolution image recovery from imageplane arrays,using convex projections[J].Journal of the Optical Society of America,1989,6(11):17151726.
[16]Irani M,Peleg S.Improving resolution by image registration[J].CVGIP,1991,53(3):231239.
[17]Elad M,Feuer A.Super resolution restoration of an image sequence:adaptive filtering approach[J].IEEE Trans Image Processing,1999,8(3):387395.
[18]Hardie R.A fast image superresolution algorithm using an adaptive Wiener filter[J].IEEE Trans Image Processing,2007,16(12):29532964.
[19]Giannis K C,Nikolaos P,G.Superresolution based on fast registration and maximum a posteriori reconstruction[J].IEEE Trans on Image Processing,2007,16(7):18211830.
[20]Marcelo V W Z,Fermin S V B,Joceji M.Determining the regularization parameters for superresolution problem[J].Signal Processing,2008,88(12):28902901.
[21]Capel D,Zisserman A.Superresolution enhancement of text image sequences[C]//Proceeding of the International Conference on Pattern Recognition.Barcelona,Spain:IEEE Computer Society,2000:600605.
[22]Farsiu S,Robinson M D.Fast and robust multiframe super resolution[J].IEEE Trans on Image Processing,2004,13(10):13271344.
[23]Li X L,Hu Y T,Gao X B,et al.A multiframe image superresolution method[J].Signal Processing,2010,90(2):405414.
[24]Ni K,Nguyen T Q.Image superresolution using support vector regression[J].IEEE Trans on Image Processing,2007,16(6):15961610.
[25]Freeman W T,Pasztor E C,Carmichael O T.Learning lowlevel vision[J].International Journal of Computer Vision,2002,40(1):2547.
[26]Yang J C,Wright J,Ma Y,et al.Image superresolution as sparse representation of raw image patches[C]//CVPR.Anchorage,AK,USA:IEEE Computer Society Press,2008:18.
[27]Yang Q X,Yang R G,Davis J,et al.Spatialdepth super resolution for range images[C]//Proceedings of Computer Vision and Pattern Recognition Minneapolis,MN,USA:IEEE Computer Society Press,2007.
[28]Dai S Y,Han M,XU W,et al.Soft edge smoothness prior for alpha channel super resolution[C]// ICCV.Brazil:IEEE Computer Society Press,2007:18.
[29]Baker S,Kanade T.Limits on superresolution and how to break them[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(9):11671183.
[30]Jian S,Jian S,Zongben X,et al.Image superresolution using gradient profile prior[C]//CVPR.Anchorage,AK,USA:IEEE Computer Society Press,2008.
[31]Glasner D,Bagon S,Irani M.Superresolution from a single image[C]// ICCV.Kyoto,Japan:IEEE Computer Society Press,2009.
[32]Yang J C,Wright J,Huang T S,et al.Image superresolution via sparse representation[J].IEEE Trans on Image Processing,2008,19(11):28612873.
[33]Chang T L,Liu T L,Chuang J H.Direct energy minimization for superresolution on nonlinear manifolds[C]// ECCV.Graz,Austria:Lecture Note in Computer Science,2006:281294.
[34]Ji H,Fermüller C.Robust waveletbased superresolution reconstruction:theory and algorithm[J].IEEE Trans on Pattern Analysis and Machine Intelligence,2009,30(4):649659.
[35]Tappen M F,Russell B C,Freeman W T.Exploiting the sparse derivative prior for superresolution and image demosaicing[C]//IEEE Workshop on Statistical and Statistical and Computational Theories of Vision.Nice,France:Lecture Note in Computer Science,2003.
[36]Freeman W T,Jones T R,Pasztor E C.Examplebased superresolution[J].IEEE Computer Graphics and Applications,2002,22(2):5665.
[37]Battiato S,Gallo G,Stanco F.Smart interpolation by anisotropic diffusion[C]// Proceedings of 12th International Conference on Image Analysis and Processing.Mantora,Italy:IEEE Computer Society Press,2003:572577.
[38]Muresan D D,Parks T W.Adaptive optimalrecovery image interpolation[J].IEEE International Conference on Acoustics,Speech,and Signal Processing,2001,3:19491952.
[39]Muresan D D,Parks T W.Adaptively quadratic (AQua) image interpolation[J].IEEE Trans on Image Processing,2004,13(5):690698.
[40]Kinebuchi K,Muresan D D,Parks T W.Image interpolation using waveletbased hidden markov trees[C]// IEEE ICASSP.Salt Lake City,UT,USA:IEEE Computer Society Press,2001.
[41]Jensen K,Anastassiou D.Subpixel edge localization and the interpolation of still images[J].IEEE Trans on Image Processing,1995,4(3):285295.
[42]Li X,Orchard M T.New edgedirected interpolation[J].IEEE Trans on Image Processing,2001,10(10):15211527.
[43]Donoho D L,Johnstone I M,Koch J C,et al.Maximum entropy and the nearly black object[J].J R Statist Soc,Series B,1992,54:4181.
[44]Donoho D L.Superresolution via sparsity constraints[J].SIAM Journal on Mathematical Analysis,1992,23(5):13091331.
[45]Zweig G.Superresolution fourier transforms by optimization and ISAR imaging[J].IEEE Proceedings Radar,Sonar,Navigation,2003,150(4):247253.
[46]张新明,沈兰荪.在小波变换域内实现图像的超分辨率复原[J].计算机学报,2003,26(9):11831189.
[47]张新明,沈兰荪.基于多尺度边缘保持正则化的超分辨率复原[J].软件学报,2003,14(6):10751081.
[48]孟庆武.预估计混叠度的MAP 超分辨率处理算法[J].软件学报,2004,15(2):207214.
[49]王程,王润生.基于MAP 框架的图像序列超分辨率和模板匹配[J].计算机学报,2003,26(8):961967.
[50]韩玉兵,吴乐南.基于自适应滤波的视频序列超分辨率重建[J].计算机学报,2006,29(4):642647.
[51]韩玉兵,束锋,孙锦涛,等.基于MGGMRES 算法的图像超分辨率重建[J].计算机学报,2007,30(6):10281034.
[52]张冬明,潘炜,陈怀新.基于MAP 框架的时空联合自适应视频序列超分辨率重建[J].自动化学报,2009,35(5):484490.
[53]宋锐,吴成柯,封颖,等.一种新的基于MAP 的纹理自适应超分辨率图像复原算法[J].电子学报,2009,37(5):11251129.
[54]田岩,王志成,柳健.图像序列的快速超分辨率恢复算法[J].电子学报,2004,32(12):20742077.
[55]汪雄良,赵侠,王正明,等.一种新的SAR超分辨成像方法[J].现代雷达,2005,27(5):14.
[56]倪伟,王正明,汪雄良.基于匹配追踪算法的SAR超分辨成像[J].宇航学报,2007,28(2):465469.
[57]王光新,王正明.SAR图像目标超分辨的变范数正则化算法[J].电子学报,2008,36,(12):23892393.
[58]董臻,朱国富,梁甸农.基于外推的SAR图像分辨率增强算法[J].电子学报,2002,30(3):259362.
[59]浦利.基于混合双立方法的MPMAP超分辨力图像插值处理算法[J].北京理工大学学报,2007,27(2):161165.
[60]范晋祥,张渊.新概念军用红外成像系统的发展[J].红外与激光工程,2008,37(3),386390.
[61]Mallat S,Hwang W L.Singularity detection and processing with wavelets[J].IEEE Trans on Information Theory,1992,32(2):617643.

备注/Memo

备注/Memo:
收稿日期:20101125 基金项目:国家重点基础研究发展计划(973)项目(2007CB311005);国防基础科研计划项目(B1420110155);福建省自然科学基金项目(A0710020);中央高校基本科研业务费专项资金(2010121066, 2010121067) *通信作者:chli@xmu.edu.cn
更新日期/Last Update: 2011-03-20