消隐点共线约束逐点畸变校正算法

赵振庆; 叶东; 吴斌; 陈刚; 杨子军

doi:10.3788/OPE.20152304.1196

您当前的位置：

首页 >

文章列表页 >

消隐点共线约束逐点畸变校正算法

信息科学 | 更新时间：2020-08-13

- 消隐点共线约束逐点畸变校正算法
- Point wise distortion correction algorithm with vanishing point collinear constraint
- 光学精密工程 2015年23卷第4期页码：1196-1204
- 作者机构：
  
  哈尔滨工业大学电气工程及自动化学院,黑龙江哈尔滨,150001
- 作者简介：
- 基金信息：
  
  国家自然科学基金资助项目(51075095)();黑龙江省自然科学基金资助项目(No.E201045)()
- DOI：10.3788/OPE.20152304.1196
  中图分类号： TP391.4
- 收稿日期：2014-12-09，
  
  修回日期：2015-01-22，
  
  纸质出版日期：2015-04-25
- 稿件说明：
移动端阅览
赵振庆, 叶东, 吴斌等. 消隐点共线约束逐点畸变校正算法[J]. 光学精密工程, 2015,23(4): 1196-1204

ZHAO Zhen-qing, YE Dong, WU Bin etc. Point wise distortion correction algorithm with vanishing point collinear constraint[J]. Editorial Office of Optics and Precision Engineering, 2015,23(4): 1196-1204
赵振庆, 叶东, 吴斌等. 消隐点共线约束逐点畸变校正算法[J]. 光学精密工程, 2015,23(4): 1196-1204 DOI： 10.3788/OPE.20152304.1196.

ZHAO Zhen-qing, YE Dong, WU Bin etc. Point wise distortion correction algorithm with vanishing point collinear constraint[J]. Editorial Office of Optics and Precision Engineering, 2015,23(4): 1196-1204 DOI： 10.3788/OPE.20152304.1196.

摘要

对逐点图像畸变校正算法进行分析和研究。分析了基于四点共线交比不变性的逐点畸变校正算法的校正精度

指出该算法由于初始使用图像点含有误差导致计算其他图像点产生了精度改变

同时给出了计算结果误差较小时的图像点位置和相对距离的选取方法。基于上述分析

提出了基于消隐点共线约束的逐点畸变校正算法。该算法利用消隐点来提高直线拟合精度

利用共线特征来约束校正图像点精度。该算法不仅可以优化基于四点共线交比不变性算法校正的图像点

同时也可以优化初始使用的图像点

从而提高所有图像点畸变校正精度。基于MATLAB的仿真实验显示:对于400万像素的镜头

校正后图像点最大误差是初始图像点噪声的29.05倍。针对实物图像对基于四点共线交比不变性法校正的图像点

基于消隐点共线约束法校正的图像点和未校正采集图像点的交比值进行对比

结果表明本文提出算法优于四点共线交比不变性算法的结果。

Abstract

Point wise correction algorithms are researched and analyzed. The correction accuracy of a point wise distortion correction algorithm based on four-point collinear cross ratio invariability is investigated. It points out that the initial used image points which contain errors will cause the accuracy change in the calculation of other image points. Then

it gives the selection method for image point locations and relative distances when the calculation error is smaller. Based on the above

this paper proposes a point wise distortion correction algorithm based on vanishing point collinear constraint. The algorithm uses vanishing points to improve the linear fit accuracy and the collinear feature to constrain the corrected distortion accuracy. It not only optimizes the image point errors of the distortion correction algorithm based on a four-point collinear cross ratio invariance

but also the initial image point errors

so that the accuracy of distortion correction of all image points is improved ultimately. The simulation based on MATLAB shows that the maximum error of corrected image points is 29.05 times that of the initial image point noise for four million pixels. In contrast to the physical image using the cross ratios from the correction algorithms based on four-point collinear invariant cross ratio and vanishing point collinear constraint as well the uncorrected image

it indicates that the performance of proposed algorithm is better than that of the algorithm based on four-point collinear cross ratio invariance.

关键词

Keywords

references

ZHANG Z Y. A flexible new technique for camera calibration [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(11): 1330-1334.

胡浩, 梁晋, 唐正宗, 等. 显微立体视觉小尺度测量系统的标定[J]. 光学精密工程, 2014, 22(8): 1985-1994. HU H, LIANG J, TANG ZH Z, et al.. Calibration of stereo microscope measurement sysems [J]. Opt. Precision Eng., 2014, 22(8): 1985-1994.

刘书桂, 姜珍珠, 董英华, 等. 采用移动光靶标的分区域相机标定[J]. 光学精密工程, 2014, 22(2): 259-265. LIU SH G, JIANG ZH ZH, DONG Y H, et al.. Sub-regional camera calibration based on moving light target [J]. Opt. Precision Eng., 2014, 22(2): 259-265.

薛俊鹏苏显渝. 基于两个正交一维物体的单幅图像相机标定[J]. 光学学报, 2012, 32(1): 0115001. XUE J P, SU X Y.Camera calibration with single image based on two orthogonal one-dimensional objects[J]. Acta Optica Sinica, 2012, 32(1): 0115001.

贺俊吉, 张广军, 杨宪铭. 基于交比不变性的镜头畸变参数标定方法[J]. 仪器仪表学报, 2004, 25(5): 597-599. HE J J, ZHANG G J, YANG X M. Approach for calibraton of lens distortion based on cross ration invariability[J].Chinese Journal of Scientific Instrument, 2004, 25(5): 597-599

唐巍, 叶东, 袁锋, 等. 视觉测量系统中一种独立的畸变校正算法[J]. 光电子·激光, 2013, 24(2): 308-315. TANG W, YE D, YUAN F, et al.. Independent distortion correction algorithm in vision measurement [J]. Journal of Optoelectronics·Laser, 2013, 24(2): 308-315.

周富强, 胡坤, 张广军. 基于共线特征点的摄像机镜头畸变校正[J]. 机械工程学报, 2006, 42(9):174-177. ZHOU F Q, HU K, ZHANG G J. Correcting distortion of camera lens with collinear points [J]. Chinese Journal of Mechanical Engineering, 2006, 42(9):174-177.

周富强, 蔡斐华. 基于非量测畸变校正的摄像机标定方法[J]. 机械工程学报, 2009, 45(8): 228-232. ZHOU F Q, CAI F H. Camera calibration method based on non-metric distortion correction [J]. Chinese Journal of Mechanical Engineering, 2009, 45(8): 228-232.

张靖, 朱大勇, 张志勇. 摄像机镜头畸变的一种非量测校正方法[J]. 光学学报, 2008, 28(8):1552-1557. ZHANG J, ZHU D Y, ZHANG ZH Y. Non-metric calibration of camera lens distortion [J]. Acta Optica Sinica, 2008, 28(8):1552-1557.

张之江, 车仁生, 黄庆成, 等. 测头成像视觉坐标测量系统中特征点成像中心获取[J]. 光学精密工程, 1998, 6(5): 12-18. ZHANG ZH J, CHEN R SH, HUANG Q CH, et al.. Centroid of characteristic point image obtain in probe imaging vision coordinate measuring system [J]. Opt. Precision Eng., 1998, 6(5): 12-18.

杨博文, 张丽艳, 叶南, 等. 装配位姿视觉检测系统中红外靶点质心定位[J]. 中国机械工程, 2014, 25(7): 965-971. YANG B W, ZHANG L Y, YE N, et al.. Centroid location algorithms of infrared target in computer-vision-based assembly pose detection system [J]. China Mechanical Engineering, 2014, 25(7):965-971.

冯新星, 张丽艳, 叶南, 等. 二维高斯分布光斑中心快速提取算法研究[J]. 光学学报, 2012, 32(5): 0512002. FENG X X, ZHANG L Y, YE N. Fast algorithms on center location of two dimensional Gaussian distribution spot [J]. Acta Optica Sinica, 2012, 32(5): 0512002.

黄风山, 钱惠芬. 三坐标测量机驱动的摄像机标定技术[J]. 光学精密工程, 2010, 18(4): 952-957. HUANG F SH, QIAN H F. Camera calibration technology driven by three-coordinate measuring machine [J]. Opt. Precision Eng., 2010, 18(4): 952-957.

HARTLEY R I, ZISSERMAN A. Multiple View Geometry in Computer Vision[M]. Cambridge:Cambridge University Press, 2000:568-582.

浏览量

292

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

任意位姿圆柱曲面透视投影失真的图像校正

LightDiffu-DCE：基于光照强度扩散的低光照图像增强

多级图特征融合引导相机位姿回归

基于机载的红外动态目标视频实时超分辨率重建

基于知识蒸馏的Transformer视觉跟踪器