基于线性特征的鱼眼图像校正方法

计算机科学 ›› 2014, Vol. 41 ›› Issue (Z6): 187-190.

基于线性特征的鱼眼图像校正方法

郭雄飞,鲁斌,李庆,薛晨阳

中北大学电子与计算机科学技术学院太原030051;中科院微电子所昆山分所昆山215300;中科院微电子所昆山分所昆山215300;中科院微电子所昆山分所昆山215300;中北大学电子与计算机科学技术学院太原030051

出版日期:2018-11-14 发布日期:2018-11-14
基金资助:
本文受汽车移动物联网总体技术研究基金项目(2012ZX03005018),智能交通创新服务系统项目(XDA06040300)资助

Calibration Method Based on Linear Feature for Fisheye Image

GUO Xiong-fei,LU Bin,LI Qing and XUE Chen-yang

Online:2018-11-14 Published:2018-11-14

摘要/Abstract

摘要： 针对鱼眼镜头采集的畸变图像,在棋盘格标定模板的基础上,提出一种有效的校正方法。该方法采用标准畸变模型,首先基于畸变图像的线性特征提出目标函数；然后利用SUSAN改进算法精确定位棋盘格角点坐标；对已得到的有限角点坐标进行畸变曲线拟合,在畸变曲线上提取所需数量的点；最后用渐近寻优算法计算最优畸变系数,使得目标函数最小。实验结果表明,该方法角点定位精确、算法时间复杂度低,能够有效校正鱼眼镜头造成的图像畸变。

关键词: 畸变校正,曲线拟合,线性特征,渐近寻优中图法分类号TP391.4文献标识码A

Abstract: In order to calibrate the distorted image captured by fisheye cameras,this paper drawed out a effective calibration method with the chessboard．Adopting standard distortion model,it firstly gets objective function on the basis of the linear feature,then detects accurately chessboard corners with a improved SUSAN algorithm．The next step it conducts curve fitting based on the previous corners．Finally the optimal distorted coefficients can be computed with progressive optimization algorithm so that objective function is minimal．Experiment results show that this method gets accurate corners coordinates,and has a low time complexity,moreover can calibrate effectively the distorted image because of fisheye lens.

Key words: Distortion calibration,Curve fitting,Linear feature,Progressive optimization

郭雄飞,鲁斌,李庆,薛晨阳. 基于线性特征的鱼眼图像校正方法[J]. 计算机科学, 2014, 41(Z6): 187-190. https://doi.org/

GUO Xiong-fei,LU Bin,LI Qing and XUE Chen-yang. Calibration Method Based on Linear Feature for Fisheye Image[J]. Computer Science, 2014, 41(Z6): 187-190. https://doi.org/

参考文献

[1] Scaramuzza D,Martinelli A,Siegwart R．A Flexible Technique for Accurate Omnidirectional Ca-mera Calibration and Structure from Motion[M].IEEE,2006
[2] Zhang Z Y．A Flexible Technique for Camera Calibration[J].IEEE Transactions on Pattern Analysis and MachineIntelligence,2000,2(11):1330-1334
[3] 郭永刚,葛庆平,冯平,等.一种基于控制点自动提取的图像畸变矫正算法[J].计算机工程与应用,2007,3(2):241-216
[4] 王占斌,赵辉,陶卫,等.广角镜头桶形畸变的样条函数修正方法[J].光电工程,2008,5(4):140-144
[5] 马广斌,章文毅,陈甫.图像几何畸变精校正研究[J].计算机工程与应用,2007,3(9):45-48
[6] Ahmed M T,Farag A A．Nonmetric calibration of camera lens distortion:differential methods and robust estimation[J]．IEEE Transactions on Image Processing,2005,4(8):1215-1230
[7] 崔洪洲,孔渊,周起勃,等.基于畸变畸变率的图像几何校正[J].应用光学,2006,7(3):183-185
[8] 韩广良,宋建忠.一种基于畸变等效曲面的图像畸变矫正[J]．光学技术,2005,1(1):122-124
[9] 郭羽,杨红,杨照金,等.CCD摄像系统镜头的畸变测量[J].应用光学,2008,9(2):179-282
[10] Slama C．Manual of photogrammetry[M]．American So-ciety of Photogrammetry,1980
[11] 徐树奎,李国辉,张军,等.基于多方向对称和匀质约束的SUSAN棋盘格角点检测算法[J].计算机科学,2011,8(9):248-252
[12] Richard I．In defence of the 8-point algorithm[C]∥Proc．International Conference on Computer Vision．1995
[13] Manolis I,Lourakis A．A Brief Description of the L-evenberg-Marquardt Algorithm Implemened by Levmar[J]．Foundation for Research and Technology,2005(2)

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed