|本期目录/Table of Contents|

[1]洪汉玉,张文莫,章秀华,等.双目视图运动图像去模糊方法[J].武汉工程大学学报,2015,37(04):45-50.[doi:10. 3969/j. issn. 1674—2869. 2015. 04. 010]
 ,SHI Yu,et al.Deblurring of double view motion image[J].Journal of Wuhan Institute of Technology,2015,37(04):45-50.[doi:10. 3969/j. issn. 1674—2869. 2015. 04. 010]
点击复制

双目视图运动图像去模糊方法(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
37
期数:
2015年04期
页码:
45-50
栏目:
出版日期:
2015-04-30

文章信息/Info

Title:
Deblurring of double view motion image
文章编号:
1674—2869(2015) 04—0045—06
作者:
洪汉玉张文莫章秀华时 愈
武汉工程大学机电工程学院,湖北 武汉 430205
Author(s):
HONG han-yu ZHANG Wen-moZHANG Xiu-huaSHI Yu
School of Mechanical and Electrical Engineering, Wuhan Institute of Technology, Wuhan 430205, China
关键词:
运动去模糊双目视图张正友标定对极几何点扩散函数路径
Keywords:
motion deblurring binocular image Zhengyou Zhang calibration the epipolar geometry point spread function path
分类号:
TB35
DOI:
10. 3969/j. issn. 1674—2869. 2015. 04. 010
文献标志码:
A
摘要:
现代工业生产线上采集到的产品图像存在运动模糊,对后期探测产品表面信息以及重建产品轮廓产生影响. 为了解决这一问题,提出一种基于双目视图的运动图像去模糊方法. 通过双目相机之间的对极几何关系以及相机模型,推导出两视点图像点扩散函数路径之间的内在联系,并运用像机标定模块求取标定矩阵中的参数. 通过此内在约束条件求取精确的点扩散函数,并对双目视点图像进行统一去模糊. 结果表明,通过双视点图像几何关系求取的点扩散函数路径来进行去模糊取得了较好的效果.
Abstract:
The images acquired from the actual production line in modern industrial production are motion-blurred, which affects the detection of the surface information and the reconstruction of the profile. To solve this problem, we proposed a deblurring method based on double view motion image. We deduced the relationship of the point spread function path between the double view images through the epipolar geometry and the camera model,and solved the parameters in the calibration matrix with the calibration system based on MATLAB, then we got the association point spread function(PSF) path based on the constraint condition,and deblurred the motion images. The experimental results show that the motion image deblurring using the PSF path solved by the geometric relationship has better effect.

参考文献/References:

[1] 洪汉玉. 现代图像图形处理与分析[M]. 武汉:中国地质大学出版社,2011:1-2.HONG Han-yu.Advanced processing and analysis for image and graphics[M]. Wuhan:China University of Geosciences press, 2005:11-12.(in Chinese)[2] 张天序. 成像自动目标识别[M]. 武汉:湖北科学技术出版社,2005:11-12.ZHANG Tian?鄄xu. The automatically recognizes of imaging[M]. Wuhan:Hubei Science and Technology Press,2005:11-12. (in Chinese)[3] 卢在盛.单幅图像运动模糊参数判定与去模糊研究[D].广西:广西大学,2011.LU Zai?鄄sheng. Motion?鄄blur indentification and delurring using single image[D]. Guangxi:Guangxi University,2011. (in Chinese)[4] 沈垣,李舜酩,毛建国,等.数字图像复原技术综述[J].中国图像图形学报,2009,14(9):1764-1775.SHEN Yuan,LI Xunmin,MAO Jianguo,et al.Digital image restoration techniques[J]. Journal of Image and Graphics,2009,14(9):1764-1775.(in Chinese)[5] 焦继超,赵保军,陶芬芳,等. 一种基于局部图像复原的天文图像增强算法[J].仪器仪表学报,2011,32(7):1455-1461.JIAO Ji?鄄chao,ZHAO Bao?鄄jun,TAO Fen?鄄fang, et al.Astronomical image enhancement algorithm based on regional image restoration[J]. Chinese Journal of Scientific Instrument,2011,32(7):1455-1461.(in Chinese)[6] 张红民,陈新平,张亚娟.一种改进的Wiener滤波图像复原算法[J].重庆理工大学学报自然科学,2010,24(7):76-80.ZHANG Hong?鄄min,CHEN Xin?鄄ping,ZHANG Ya?鄄juan.Improved wiener filter algorithm for image restoration[J].Journal of Chongqing Insitute of Technology Natural Science,2010,24(7):76-80. (in Chinese)[7] RASKAR R, AGRAWAL A, TUMBLIN J. Coded exposure photography:Motion deblurring using fluttered shutter[J]. ACM Trans Graph, 25(3):795-804.[8] SIMON M K Probability distributions involving gaussian random variables:a handbook for engineers[J].Scientists and Mathematicians Springer,2002:1378?鄄1382.[9] HONG Han-yu, ZHANG Tian-xu. Fast restoration approach for rotational motion blurred image based on deconvolution along the blurring path[J].Optical Engineering,42(12):3471-3486.[10] LEVIN A. Blind motion deblurring using image statistics[C]//In Advances in Neural Information Processing Systems NIPS .2006(2):1738-1742.[11] CHEN W G,NANDHAKUMAR N,MATIN W N. Image motion estimation from motion smear-a new computational model[J].Pattern Analysis and Machine Intelligence,1996,18(4):412-425.[12] RAV-Acha A, PELEG S. Two motion-blurred images are better than one[J]. Pattern Recognition Letters,2005,26(2-3):311-317.[13] BEN ME,Nayar S K. Motion-based motion eblurring[J]. IEEE Trans on Pattern Analysis and Machine Intelligence,2004, 26(6):689-698.[14] YUAN L, SUN J, QUAN L, et al. Shum Image deblurring with blurred/noisy image pairs[J]. ACM TOG,26(3),374-379.[15] YITZHAKY Y I, MOR A L, KOPEIKA N. Direct method for restoration of motion blurred images[J]. J Opt Soc Am A,15(6):1512-1519. [16] TSIN Y, RAMESH V, KANADE T. Statistical calibration of CCD image process[J]. Proc IEEE Int’l Conf Computer Vision,2001:480-487.[17] HAMMETT S T,GEORGESON M A,GOREA A.Motion Blur and Motion Sharpening:Temporal Smear and Local Contrast Non?鄄linearity[J]. Vision Resarch,1988(38):20-99. [18] 章秀华,洪汉玉,华夏,等.动目标多视点基于标定的去模糊方法[J].武汉工程大学学报,2014,36(12):56-62.ZHANG Xiuhua, HONG Han-yu, HUA Xia, et al. Multi-view motion of moving target deblurring based on calibration[J]. Journal of Wuhan Institute of Technology,2014,36(12):56-62.(in Chinese)

相似文献/References:

备注/Memo

备注/Memo:
收稿日期:2015-03-05基金项目:国家自然科学基金面上项目(61175013),湖北省自然科学基金创新群体项目(2012FFA046);武汉市国际科技合作计划项目(编号2014030709020310).作者简介:洪汉玉(1964-),男,湖北阳新人,教授,博士,研究方向:图像处理与模式识别.
更新日期/Last Update: 2015-05-25