浏览全部资源
扫码关注微信
北京工业大学 信息学部, 北京 100124
[ "彭天奇(1995-),男,山东菏泽人,硕士研究生。2018年于青岛农业大学获得学士学位。主要研究方向为图像处理、模式识别。E-mail: ptq17812103095@163.com" ]
[ "肖创柏(1962-),男,湖南临湘人,教授、博士生导师。1983年于湘潭大学获得学士学位,1986年于西北工业大学获得硕士学位,1995年于清华大学获得博士学位。主要研究方向为数字信号处理,音视频信号处理与网络通信。E-mail:cbxiao@bjut.edu.cn" ]
收稿日期:2020-08-27,
修回日期:2020-10-14,
纸质出版日期:2021-02-15
移动端阅览
彭天奇,禹晶,郭乐宁等.基于跨尺度字典学习的图像盲解卷积算法[J].光学精密工程,2021,29(02):338-348.
PENG Tian-qi,YU Jing,GUO Le-ning,et al.Blind image deconvolution via cross-scale dictionary learning[J].Optics and Precision Engineering,2021,29(02):338-348.
彭天奇,禹晶,郭乐宁等.基于跨尺度字典学习的图像盲解卷积算法[J].光学精密工程,2021,29(02):338-348. DOI: 10.37188/OPE.20212902.0338.
PENG Tian-qi,YU Jing,GUO Le-ning,et al.Blind image deconvolution via cross-scale dictionary learning[J].Optics and Precision Engineering,2021,29(02):338-348. DOI: 10.37188/OPE.20212902.0338.
在模糊核未知情况下利用模糊图像对清晰图像进行复原称为图像盲解卷积问题,这是一个欠定逆问题,现有的大部分算法通过引入模糊核和清晰图像的先验知识来约束问题的解空间。本文提出了一种基于跨尺度字典学习的图像盲解卷积算法,采用降采样图像训练稀疏表示的字典,并将图像纹理区域在该字典下的稀疏表示作为正则化约束引入盲解卷积目标函数中。图像降采样过程减弱了图像的模糊程度,且图像中存在冗余的跨尺度相似块,利用更清晰的图像块训练字典能够更好地对清晰图像进行稀疏表示,减小稀疏表示误差;同时,由于在纹理区域清晰图像的稀疏表示误差小于模糊图像的稀疏表示误差,在该字典下对图像中的纹理块进行稀疏表示,使重建图像偏向清晰图像。本文的算法在Kohler数据集上复原结果的平均峰值信噪比为29.54 dB。在大量模糊图像上的实验验证了本文的算法能够有效解决大尺寸模糊核的复原,并具有良好的鲁棒性。
Blind image deconvolution recovers a sharp image from a blurred image when the blur kernel is unknown. To solve this underdetermined inverse problem, most existing methods exploit various image priors to constrain the solution. In this study, we propose a blind deconvolution method based on cross-scale dictionary learning, in which the down-sampled blurry image is used to learn a dictionary as training samples and the texture region is represented sparsely over the dictionary as the regularization term. Because the down-sampling process weakens the blur of the image, it will result in the formation of redundant cross-scale similar patches. To ensure that a sharp image is represented sparsely, sharper image patches from the down-sampled image in this study were used to learn the dictionary as training samples. The results showed that the sparse representation error of the texture patch from the sharp image was less than that from the blurred image, further diminishing the sparse representation error over the dictionary, and the intermediate latent image approached the sharp image. The mean peak signal-to-noise ratio of the results by our method on the dataset of Kohler et al. is 29.54 dB. Experimental results on blurry images demonstrated that our method can estimate large blur kernels accurately and that it has good robustness.
王宗跃 , 夏启明 , 蔡国榕 , 等 . 自适应图像组的稀疏正则化图像复原 [J]. 光学 精密工程 , 2019 , 27 ( 12 ): 2713 - 2721 .
WANG Z Y , XIA Q M , CAI G R , et al . . Image restoration based on adaptive group images sparse regularization [J]. Opt. Precision Eng. , 2019 , 27 ( 12 ): 2713 - 2721 . (in Chinese)
杨平先 , 陈明举 . 一种基于L1范数的非局部变分图像复原模型 [J]. 液晶与显示 , 2017 , 32 ( 8 ): 635 - 641 .
YANG P X , CHEN M J . A non local total variation based on L1 norm for image recovery [J]. Chinese Journal of Liquid Crystals and Displays , 2017 , 32 ( 8 ): 635 - 641 . (in Chinese)
武兴睿 . 基于PSF重构和改进的最大后验估计的自适应光学图像复原算法 [J]. 液晶与显示 , 2019 , 34 ( 9 ): 921 - 927 .
WU X R . Adaptive optical image restoration method based on PSF reconstruction and improved Maximum A Posteriori estimation [J]. Chinese Journal of Liquid Crystals and Displays , 2019 , 34 ( 9 ): 921 - 927 . (in Chinese)
CHAN T , WONG C . Total variation blind deconvolution [J]. IEEE Transactions on Image Processing , 1998 , 7 ( 3 ): 370 - 375 .
LEVIN A , WEISS Y , DURAND F , et al .. Efficient marginal likelihood optimization in blind deconvolution [C]. IEEE Conference on Computer Vision and Pattern Recognition. Providence , RI. IEEE , 2011 : 2657 - 2664 .
KRISHNAN D , TAY T , FERGUS R . Blind deconvolution using a normalized sparsity measure [C]. IEEE International Conference on Computer Vision. Providence , RI USA. IEEE , 2011 : 233 - 240 .
ANGER J , FACCIOLO G , DELBRACIO M . Blind image deblurring using the I0 gradient prior [J]. Image Processing On Line , 2019 , 9 : 124 - 142 .
MICHAELI T , IRANI M . Blind deblurring using internal patch recurrence [C]. Springer European Conference on Computer Vision. Zurich , Switzerland. Springer , 2014 : 783 - 798 .
PAN J , SUN D , PFISTER H , et al .. Blind image deblurring using dark channel prior [C]. IEEE Conference on Computer Vision and Pattern Recognition . LasVegas , NV , USA. IEEE , 2016 : 1628 - 1636 .
YAN Y , REN W , GUO Y , et al .. Image deblurring via extreme channels prior [C]. IEEE Conference on Computer Vision and Pattern Recognition. Losangeles , CA , USA. IEEE , 2016 : 6978 - 6986 .
CHEN L , FANG F , WANG T , et al .. Blind image deblurring with local maximum gradient prior [C]. IEEE Conference on Computer Vision and Pattern Recognition. Honolulu , HI , USA. IEEE , 2019 : 1742 - 1750 .
HU Z , HUANG J , YANG M . Single image deblurring with adaptive dictionary learning [C]. IEEE International Conference on Image Processing , Hong Kong. IEEE , 2010 : 1169 - 1172 .
ZHANG H , YANG J , ZHANG Y , et al .. Close the loop: Joint blind image restoration and recognition with sparse representation prior [C]. IEEE International Conference on Computer Vision. Barcelona , Spain. IEEE , 2011 : 770 - 777 .
AHARON M , ELAD M , BRUCKSTEIN A . K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation [J]. IEEE Transactions on Signal Processing , 2006 , 54 ( 11 ): 4311 - 4322
常振春 , 禹晶 , 肖创柏 , 等 . 基于稀疏表示和结构自相似性的单幅图像盲解卷积算法 [J]. 自动化学报 , 2017 , 43 ( 11 ): 1908 - 1919 .
CHANG ZH CH , YU J , XIAO CH B , et al . . Single image blind deconvolution using sparse representation and structural self-similarity [J]. Acta Automatica Sinica , 2017 , 43 ( 11 ): 1908 - 1919 . (in Chinese)
GlASNER D , BAGON S , IRANI M . Super-resolution from a single image [C]. IEEE International Conference on Computer Vision. Kyoto. IEEE , 2009 : 349 - 356 .
XU L , JIA J . Two-phase kernel estimation for robust motion deblurring [C]. Springer European Conference on Computer Vision. Berlin Heidelberg. Springer , 2010 : 157 - 170 .
SUN L , CHO S , WANG J , HAYS J . Edge-based blur kernel estimation using patch priors .[C] IEEE International Conference on Computational Photography . Cambridge , MA. IEEE , 2013 : 1 - 8 .
KOHLER R , HIRSCH M , MOHLER B . Recording and playback of camera shake: Benchmarking blind deconvolution with a real-world database [C]. Springer European Conference on Computer Vision. Florence , Italy. Springer , 2012 : 27 - 40 .
0
浏览量
870
下载量
2
CSCD
关联资源
相关文章
相关作者
相关机构