[1]王伟,郑津津,刘星,等.图像复原中的模糊参数估计[J].智能系统学报,2012,7(04):315-320.
 WANG Wei,ZHENG Jinjin,LIU Xing,et al.Estimation of blur parameters in image restoration[J].CAAI Transactions on Intelligent Systems,2012,7(04):315-320.
点击复制

图像复原中的模糊参数估计(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第7卷
期数:
2012年04期
页码:
315-320
栏目:
出版日期:
2012-08-25

文章信息/Info

Title:
Estimation of blur parameters in image restoration
文章编号:
1673-4785(2012)04-0315-06
作者:
王伟1郑津津1刘星1周洪军2沈连婠1
1.中国科学技术大学 精密机械与精密仪器系,安徽 合肥 230027;
2.中国科学技术大学 国家同步辐射实验室,安徽 合肥 230029
Author(s):
WANG Wei1 ZHENG Jinjin1 LIU Xing1 ZHOU Hongjun2 SHEN Lianguan1
1.Department of Precision Machinary and Precision Instrumentation, University of Science and Technology of China, Hefei 230027, China;
2.National Synchrotron Radiation Laboratory, University of Science and Technology of China, Hefei 230029, China
关键词:
运动模糊Hough变换〖KG-*1/3〗-〖KG-*1/5〗参数曲线圆心拟合极坐标变换
Keywords:
motion blur Hough transformparameter curve center fitting polar transformation
分类号:
TP18
文献标志码:
A
摘要:
图像复原技术在航空拍摄和机器视觉中是提高图像质量的重要手段.针对有相对运动情况的图像存在的降质模糊,分析了运动模糊的数学原理,针对不同运动模糊采取了不同去模糊方式.即先判断图像存在的运动模糊方式,针对直线运动通过Hough变换法和误差参数法相结合的方法估计降质参数,而对于旋转运动则采用曲线拟合和极坐标转换相结合的方法来估计降质参数.通过维纳滤波方法复原图像,提高图像质量.利用提出的自适应方法对直线运动模糊(参数为(30,70°))和旋转运动模糊(参数为(128,128),20°)分别作了实验计算.对比实验表明,这种由粗及精的方法能准确估计模糊参数,与传统处理方法相比,更加便捷有效.
Abstract:
The image restoration technique is an important way to improve the quality of images, especially in flight photography and machine vision. In this paper, the mathematic principle was analyzed first, and then different deblurring methods were employed correspondingly for each motion condition; specifically, the type of motion blur was judged. For linear motion blur, both the Hough transform and errorparameter were taken to estimate parameters. Furthermore, for rotational motion blur, the curvefitting and polar transform were combined to estimate degraded parameters. Both were followed by Wiener filtering to recover images and improve image vision quality. The proposed adaptive algorithm was used to calculate the linear motion blur parameter (30,70°) and rotational motion blur parameter ((128,128), 20°). The results show that the algorithm can estimate blur parameters accurately. Compared with traditional handling, this proposed method has the advantages of high precision, robustness, and speed.

参考文献/References:

[1]WIENER N. Extrapolation, interpolation, and smoothing of stationary time series[M]. New York: Wiley, 1949: 215.
[2]WILLIAM H R. Bayesianbased iterative method of image restoration[J]. Journal of the Optical Society of America, 1972, 62 (1): 5559.
[3]RAVACHA A, ANDPELEG S. Two motion blurred images are better than one[J]. Pattern Recognition Letters, 2005, 26: 311317. 
[4]YUAN L, SUN J, QUAN L, et al. Image deblurring with blurred noisy image pairs[J]. ACM Trans Graphics, 2007, 26 (3): 110 .
[5]BENEZRA M, ANDNAYAR, S. K. Motionbased motion deblurring[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2004, 26(6): 689698. 
[6]RASKAR R, AGRAWAL A, ANDTUMBLIN J. Coded exposure photography: motion deblurring using fluttered shutter[J]. ACM Transactions on Graphics, 2006, 25(3): 95804.
[7]HUNT B R, CANNON T M. Nonstationary assumptions of Gaussian models of images[J]. IEEE Transactions on Systems, Man, and Cybern, 1976, 2: 876882. 
[8]YITZHAKY Y, KOPEIKA N S. Identification of blur parameters from motion blurred images[J]. Graphical Models and Image Processing, 1997, 59(5): 321332. 
[9]JACK E. BRESENHAM. Algorithm for computer control of〖LL〗a digital plotter[J]. IBM Systems Journal, 1965, 4(1): 2530.
[10]DUDA R O, HART P E. Use of the hough transformation to detect lines and curves in pictures[J]. Communications of the ACM, 1972, 15: 1115.
[11]CANNY J. A computational approach to edge detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, 8(6): 679698.
[12]邹谋炎. 反卷积和信号复原[M]. 北京:国防工业出版社, 2001: 200202.
[13]ALDRICH J. Doing least squares: perspectives from Gauss and Yule[J]. International Statistical Review, 1998,66 (1): 6181.

备注/Memo

备注/Memo:
收稿日期: 2011-11-17.
网络出版日期:2012-07-12.
基金项目:国家自然科学基金资助项目(10575097,10775128,61073109);高校博士点基金资助项目(20060358050);国家自然基金大科学装置联合基金资助项目 (10979065);“111”工程资助项目(B07033).
通信作者:王伟.
E-mail:wwei009@mail.ustc.edu.cn.
作者简介:
王伟,男,1987年生,博士研究生,主要研究方向为图像处理与算法设计.
郑津津,男,1963年生,博士生导师,教授,主要研究方向为计算机辅助微加工误差仿真、细分曲面造型方法理论、基于图像的曲面重构、基于散乱数据的曲面重构、太阳光帆仿真理论、基于多边形曲面的曲面造型与纹理映射等.
刘星,男,1988年生,硕士研究生,主要研究方向为图像视频分析.
更新日期/Last Update: 2012-09-26