|本期目录/Table of Contents|

[1]洪汉玉,章秀华,华夏,等.动目标多视点图像基于标定的去模糊方法[J].武汉工程大学学报,2014,(12):6.
　Hanyu Hong,Xiuhua Zhang,Xia Hua,et al. Multi-view Motion Deblurring Method Based on? Calibration for Moving Target[J].Journal of Wuhan Institute of Technology,2014,(12):6.
点击复制

动目标多视点图像基于标定的去模糊方法(/HTML)
分享到：

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

卷:
期数:
2014年12期
页码:
6
栏目:
机电与信息工程
出版日期:
2014-12-31

文章信息/Info

Title:

Multi-view Motion Deblurring Method Based on? Calibration for Moving Target

作者:
洪汉玉; 章秀华; 华夏; 白浩玉
武汉工程大学图像处理与智能控制研究室 湖北 武汉 430205
Author(s):

Hanyu Hong; Xiuhua Zhang; Xia Hua; Haoyu Bai

Laboratory of Image Processing and Intelligent Control, Wuhan Institute of Technology, Wuhan 430205
关键词:
点扩散函数; 多视点; 运动模糊; 图像复原
Keywords:
point spread function;  multi – view;  motion blur;  image restoration
分类号:
-
DOI:
-
文献标志码:
A
摘要:
三维视觉检测技术在工业生产线上有着远大的应用前景，由于生产线动目标多视点图像存在模糊问题，然而现有的图像复原技术存在着局限性，使得复原后的图像很难用来进行三维重建，进而也不能实现三维视觉检测。针对这个问题，本文提出了一种新的动目标多视点图像去模糊方法。首先估计出部分视点模糊图像(本文以两个视点的模糊图像为例)点扩展函数。其次根据多视点几何理论，利用相机标定方法，得到各个相机的参数矩阵。结合估计出的部分视点的点扩展函数和相机参数矩阵，以直线运动为例，建立多视点模糊图像的点扩展函数关系。最后在确保各视点模糊路径对应的关系下，对多视点图像进行整体去模糊。一系列实验验证了本文提出的动目标多视点图像去模糊方法的有效性。
Abstract:
The three-dimensional vision inspection is an important technology to achieve automatic and intelligence in the field of manufacturing industry. But in the practical production line, a motion blur image will be obtained by the camera. It is difficult to use these motion blur images for three-dimensional reconstruction and it is not possible to realize the visual inspection. The proposed method not only improves the clarity of motion-blurred images from different angles, but also ensures the corresponding relationship between these point spread functions of the motion-blurred images from different angles. Firstly, the point spread function of some view motion-blurred images is estimated (We take two view motion-blurred images as an example). Then, the parameter matrix of each camera is obtained by a camera calibration method based on the multi-view geometry theory. The corresponding relationship between these point spread functions is built by using the parameter matrix of each camera. Finally, multi-view motion blurred images are restored as a whole. A series of experiments show the effectiveness of the proposed method.

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.

[2] Chen W G, Nandhakumar N, Martin W N. Image motion estimation from motion smear-a new computational model[J].Pattern Analysis and Machine Intelligence,1996,18(4):412-425.

[3] Hanyu Hong, Tianxu Zhang. Fast restoration approach for rotational motion blurred image based on deconvolution along the blurring paths. Optical Engineering, 2003, 42(12): 3471-3486.



[4] Hong Hanyu, Li Liangcheng, Zhang Tianxu. Blind restoration of real turbulence-degraded image with complicated backgrounds using anisotropic regularization[J]. OpticsCommunications,2012,285(2): 4977-4986.

[5] 马颂德，张正友.计算机视觉--计算机理论与算法基础[M].北京: 科学出版社,1998.

[6] Zhang Zhengyou. A flexible new technique for camera calibration [C].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,2(11):1330-1334.

[7] Hong Hanyu , Park I K. Single-image motion deblurring using adaptive anisotropic regularization[J].Optical Engineering,2010,49(9):1-13.

[8] HONG Han-Yu; ZHANG Tian-Yu; Yu Guo-Liang. Regularized restoration algorithm of astronautcal turbulence-degraded images using maximum-likelihood estimation, Journal of Infrared and Millimeter Waves，2005, 24(2):130-134.





相似文献/References:

[1]付　航,章秀华*,贺　武.多视点去模糊算法在双核ＤＳＰ上的并行优化[J].武汉工程大学学报,2015,37(11):74.[doi:10. 3969/j. issn. 1674-2869. 2015. 1１. 015]
　,Ｐａｒａｌｌｅｌ ｏｐｔｉｍｉｚａｔｉｏｎ ｏｆ ｍｕｌｔｉ－ｖｉｅｗ ｄｅｂｌｕｒｒｉｎｇ ａｌｇｏｒｉｔｈｍ ｉｎ ｄｕａｌ－ｃｏｒｅ ｄｉｇｉｔａｌ ｓｉｇｎａｌ ｐｒｏｃｅｓｓｏｒ[J].Journal of Wuhan Institute of Technology,2015,37(12):74.[doi:10. 3969/j. issn. 1674-2869. 2015. 1１. 015]
[2]章秀华,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,(12):56.[doi:0.3969/j.issn.1674-2869.2014.012.011]
　G,,et al.Ｍｕｌｔｉ－ｖｉｅｗ ｍｏｔｉｏｎ ｏｆ ｍｏｖｉｎｇ ｔａｒｇｅｔ ｄｅｂｌｕｒｒｉｎｇ ｂａｓｅｄ ｏｎ ｃａｌｉｂｒａｔｉｏｎ[J].Journal of Wuhan Institute of Technology,2014,(12):56.[doi:0.3969/j.issn.1674-2869.2014.012.011]

备注/Memo

备注/Memo:
-
更新日期/Last Update: 2014-11-27