两次引导滤波的显微视觉散焦图像快速盲复原
|
摘要
针对显微图像盲复原算法存在的计算量大、振铃效应以及噪声敏感的问题,提出贝叶斯框架下两次引导滤波的快速盲复原算法。利用显微图像成像原理中基于深度信息估计点扩展函数的概率模型,构建了贝叶斯框架下盲复原的最小优化问题;通过分析最大后验概率的最小优化问题求解过程,推出了实施引导滤波器可快速求解优化问题的结论;为有效去除振铃和噪声,设计了两次引导滤波的求解方案,其将第一次引导滤波求解的结果作为优化问题的二次输入。实验结果表明,复原结果的像素误差率约为0.04,较常用盲复原算法的复原准确度提高了约20%,运行时间也大幅缩短,该方法能有效应用于显微视觉下微装配散焦图像盲复原的工程实践中。
To solve the problems of large computation cost,ringing and noise sensitivity in blind restoration algorithms for microscopic images,the blind restoration algorithm under Bayesian framework based on two guided filterings is proposed.The depth information of microscopic image is used to estimate the probabilistic model of point spread function,and a minimum optimization problem under the Bayesian framework is built.The guided filtering is applied to searching the optimal solution through analyzing the solving scheme of the minimum optimization problem of the maximum posterior probability.The solution scheme of the two guided filtering algorithms is designed for removing ringing and noise,which means the restoration result of the first guided filtering will serve as input of the optimization problem again.Experimental results show that the pixel error rate of recovery result is around 0.04,which increases by 20% compared to those of other commonly used algorithms,and the running time is significantly shortened.The proposed algorithm can be used in assembly of the micro-structures for defocused image blind restoration.
To solve the problems of large computation cost,ringing and noise sensitivity in blind restoration algorithms for microscopic images,the blind restoration algorithm under Bayesian framework based on two guided filterings is proposed.The depth information of microscopic image is used to estimate the probabilistic model of point spread function,and a minimum optimization problem under the Bayesian framework is built.The guided filtering is applied to searching the optimal solution through analyzing the solving scheme of the minimum optimization problem of the maximum posterior probability.The solution scheme of the two guided filtering algorithms is designed for removing ringing and noise,which means the restoration result of the first guided filtering will serve as input of the optimization problem again.Experimental results show that the pixel error rate of recovery result is around 0.04,which increases by 20% compared to those of other commonly used algorithms,and the running time is significantly shortened.The proposed algorithm can be used in assembly of the micro-structures for defocused image blind restoration.
引文
[1]Cao Lei,Chen Hongbin,Qiu Qi,et al.Blind image deconvolution based on power law distribution applied in optoelectronic detections system[J].Chinese J Lasers,2015,42(3):0308007.曹雷,陈洪斌,邱琪,等.基于指数律分布的快速盲图像解卷积在光电探测系统中的应用[J].中国激光,2015,42(3):0308007.
[2]Zhao Qingqing,Zhang Tao,Zheng Weibo.Research on high resolution digital refocusing of light field imaging based on microlens array[J].Laser&Optoelectronics Progress,2016,53(10):101001.赵青青,张涛,郑伟波.基于微透镜型光场成像的高分辨率数字对焦技术研究[J].激光与光电子学进展,2016,53(10):101001.
[3]Liu Xiaohui,Guo Cheng′an,Hu Jiasheng.A modified Wiener filtering for restoration of ring-coded aperture image in inertial confinement fusion[J].Acta Optica Sinica,2004,24(8):1045-1050.刘晓辉,郭成安,胡家升.惯性约束聚变中环孔编码图像恢复的改进维纳滤波方法[J].光学学报,2004,24(8):1045-1050.
[4]Yoo J C,Ahn C W.Image restoration by blind-Wiener filter[J].IET Image Processing,2014,8(12):815-823.
[5]Hu Xiaoping,Chen Guoliang,Mao Zhengyu,et al.Study on Wiener filtering for restoration of defocus blur image[J].Chinese Journal of Scientific Instrument,2007,28(3):479-482.胡小平,陈国良,毛征宇,等.离焦模糊图像的维纳滤波复原研究[J].仪器仪表学报,2007,28(3):479-482.
[6]Khan M K,Morigi S,Reichel L,et al.Iterative methods of Richardson-Lucy-type for image deblurring[J].Numerical Mathematics Theory Methods&Applications,2013,6(1):262-275.
[7]Zhang H C,Wipf D,Zhang Y N.Multi-observation blind deconvolution with an adaptive sparse prior[J].IEEETransactions on Pattern Analysis and Machine Intelligence,2014,36(8):1628-1643.
[8]Song C W,Deng H,Gao H J,et al.Bayesian non-parametric gradient histogram estimation for texture-enhanced image deblurring[J].Neurocomputing,2016,197(1):95-112.
[9]Cho S Y,Lee S Y.Fast motion deblurring[J].ACM Transactions on Graphics,2009,28(5):89-97.
[10]Duan Jiangyong,Meng Gaofeng,Xiang Shiming,et al.Restoring motion blurred image based on edge region constraints[J].Joural of Computer-Aided Design&Computer Graphics,2012,24(8):1038-1046.段江永,孟高峰,向世明,等.边缘区域约束引导的运动模糊图像复原[J].计算机辅助设计与图形学学报,2012,24(8):1038-1046.
[11]Cao Y,Fang S,Wang F.Single image multi-focusing based on local blur estimation[C].IEEE International Conference on Image and graphics(ICIG),2011:168-175.
[12]Elad M.Retinex by two bilateral filters[C].International Conference on Scale-Space Theories in Computer Vision,2005:217-229.
[13]He K M,Sun J.Fast guided filter[EB/OL].(2015-05-05)[2016-02-09].https://arxiv.org/abs/1505.00996.
[14]Takeda H,Seo H J,Milanfar P.Statistical approaches to quality assessment for image restoration[C].International Conference on Consumer Electronic(ICCE),2008:1-2.
[15]Levin A,Weiss Y,Durand F,et al.Understanding and evaluating blind deconvolution algorithms[C].IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2009:1964-1971.
[2]Zhao Qingqing,Zhang Tao,Zheng Weibo.Research on high resolution digital refocusing of light field imaging based on microlens array[J].Laser&Optoelectronics Progress,2016,53(10):101001.赵青青,张涛,郑伟波.基于微透镜型光场成像的高分辨率数字对焦技术研究[J].激光与光电子学进展,2016,53(10):101001.
[3]Liu Xiaohui,Guo Cheng′an,Hu Jiasheng.A modified Wiener filtering for restoration of ring-coded aperture image in inertial confinement fusion[J].Acta Optica Sinica,2004,24(8):1045-1050.刘晓辉,郭成安,胡家升.惯性约束聚变中环孔编码图像恢复的改进维纳滤波方法[J].光学学报,2004,24(8):1045-1050.
[4]Yoo J C,Ahn C W.Image restoration by blind-Wiener filter[J].IET Image Processing,2014,8(12):815-823.
[5]Hu Xiaoping,Chen Guoliang,Mao Zhengyu,et al.Study on Wiener filtering for restoration of defocus blur image[J].Chinese Journal of Scientific Instrument,2007,28(3):479-482.胡小平,陈国良,毛征宇,等.离焦模糊图像的维纳滤波复原研究[J].仪器仪表学报,2007,28(3):479-482.
[6]Khan M K,Morigi S,Reichel L,et al.Iterative methods of Richardson-Lucy-type for image deblurring[J].Numerical Mathematics Theory Methods&Applications,2013,6(1):262-275.
[7]Zhang H C,Wipf D,Zhang Y N.Multi-observation blind deconvolution with an adaptive sparse prior[J].IEEETransactions on Pattern Analysis and Machine Intelligence,2014,36(8):1628-1643.
[8]Song C W,Deng H,Gao H J,et al.Bayesian non-parametric gradient histogram estimation for texture-enhanced image deblurring[J].Neurocomputing,2016,197(1):95-112.
[9]Cho S Y,Lee S Y.Fast motion deblurring[J].ACM Transactions on Graphics,2009,28(5):89-97.
[10]Duan Jiangyong,Meng Gaofeng,Xiang Shiming,et al.Restoring motion blurred image based on edge region constraints[J].Joural of Computer-Aided Design&Computer Graphics,2012,24(8):1038-1046.段江永,孟高峰,向世明,等.边缘区域约束引导的运动模糊图像复原[J].计算机辅助设计与图形学学报,2012,24(8):1038-1046.
[11]Cao Y,Fang S,Wang F.Single image multi-focusing based on local blur estimation[C].IEEE International Conference on Image and graphics(ICIG),2011:168-175.
[12]Elad M.Retinex by two bilateral filters[C].International Conference on Scale-Space Theories in Computer Vision,2005:217-229.
[13]He K M,Sun J.Fast guided filter[EB/OL].(2015-05-05)[2016-02-09].https://arxiv.org/abs/1505.00996.
[14]Takeda H,Seo H J,Milanfar P.Statistical approaches to quality assessment for image restoration[C].International Conference on Consumer Electronic(ICCE),2008:1-2.
[15]Levin A,Weiss Y,Durand F,et al.Understanding and evaluating blind deconvolution algorithms[C].IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2009:1964-1971.