基于两平行线及其线上三点的摄像机标定方法

贺科学; 李树涛; 胡建文

光学技术, 2016, 42 (6): 545, 网络出版: 2016-12-23

基于两平行线及其线上三点的摄像机标定方法

Camera calibration based on two parallel lines and three collinear points

论文大纲

贺科学 ^1,2,*李树涛 ¹胡建文 ²

作者单位

¹ 湖南大学电气与信息工程学院, 湖南长沙 410082

² 长沙理工大学电气与信息工程学院, 湖南长沙 410076

光学测量摄像机标定智能交通系统交通标线 optical measurement camera calibration intelligent transportation system road lane markings

摘要

在大视场交通监控摄像机的标定中, 由于不便于放置大尺寸标定靶标, 常采用交通标线构成的特定几何图形替代标定靶标, 但许多交通场景仅有一组平行的车道分界交通标线, 难以获得传统标定方法所需的标定约束条件。针对此类交通场景, 选取两平行交通标线和其中一条线上的三点作为标定参考物, 依靠其中一条线上三点在单幅图像中的像素坐标、三点间的两个距离值及平行线间距作为约束条件来完成摄像机的标定。建立了图像坐标系与世界坐标系之间新的换算关系, 推导出了摄像机内外参数的求解公式, 探讨了旋转角对测量误差的影响。实验结果表明, 在约束条件不足的情况下能够获得较好的摄像机参数近似解, 沿道路方向上的测距精度优于对比文献的结果, 具有标定参考物选取容易、操作简便、几何约束条件少等优点, 适用于交通监控系统中大视场摄像机的快速标定场合。

Abstract

Due to the characteristics of the traffic scenes, it is difficult to place calibration patterns in order to calibrate the roadside camera. Traffic lane marks are usually substituted for calibration patterns. However, there is only a set of parallel lane dividing lines in many traffic scenes. Calibration objects are not enough available in the traditional camera calibration methods. In the condition of insufficient constraints, two parallel lines and three collinear points on road lanes are used as calibration reference objects. The constraint conditions consist of the spacing of two parallel lines, two lengths between three collinear points and their corresponding coordinates in an image. The formulas are derived to solve camera parameters. The relationship between pixels in traffic image and points in the world coordinate system is built. The effect caused by ignoring swing angle is reassessed. Experimental results show that approximate solutions of camera parameters are satisfying despite the lack of sufficient constraints. The measurement precision in the direction of the road is superior to Fung's camera calibration method. The advantages of the proposed method are the easy selection of the reference object, good flexibility and fewer constraints. It is very suitable for expeditious calibration of the camera in the transportation surveillance system.

参考文献

[1] DUBSKA M, et al. Fully automatic roadside camera calibration for traffic surveillance［J］. IEEE Transactions on Intelligent Transportation Systems, 2014,16(3):1162-1171.

[2] 杨宁,等. 基于小尺寸靶标组合的大视场摄像机标定方法［J］. 光电子·激光,2013,24(8):1569-1575.

YANG N, et al. Calibration method of camera with large field-of-view based on spliced small targets［J］. Journal of Optoelectronics Laser, 2013,24(8): 1569-1575.

[3] HAI Dinh,et al. Simple method for camera calibration of roundabout traffic scenes using a single circle［J］. IET Intelligent Transport Systems, 2013, 8(3): 175-182.

[4] 许益铭,等.基于直线特征的摄像机标定方法研究［J］. 光学技术,2007,33(6):841-844.

XU Y M, et al. Camera calibration method research with linear feature［J］. Optical Technique, 2007, 33(6):841-844.

[5] LUGANG Z,et al. A camera calibration method based on two orthogonal vanishing points［J］. Concurrency and Computation: Practice and Experience, 2014, 26(5): 1185-1199.

[6] 陈爱华,等. 基于正交消失点对的摄像机标定方法［J］. 仪器仪表学报,2012,33(1):161-166.

Chen A H,et al. Camera calibration method based on orthogonal vanishing point pair［J］. Chinese Journal of Scientific Instrument, 2012,33(1):161-166.

[7] ZHENG Y,et al. A practical roadside camera calibration method based on least squares optimization［J］. IEEE Transactions on Intelligent Transportation Systems, 2014, 15 (2):831-843.

[8] 王秀平,等. 基于正三角形平面模板的摄像机标定与重建［J］.光学技术,2015,41(01): 34-38.

WANG X P,et al. Camera calibration and reconstruction using a planar pattern spliced by equilateral triangle［J］. Optical Technique, 2015,41(01): 34-38.

[9] GEORGE S K F,et al. Camera calibration from road lane marking［J］. Optical Engineering,2003, 42(10): 2967-2977.

[10] 李勃,等. 路况PTZ摄像机自动标定方法［J］. 北京邮电大学学报,2009,32(4): 25-29.

LI B,et al. Automatic calibration method for PTZ camera［J］. Journal of Beijing University of Posts and Telecommunications, 2009,32(4): 25-29.

[11] MIYAGAWA I,et al. Simple camera calibration from a single image using five points on two orthogonal 1-D objects［J］. IEEE Transactions on Image Processing, 2010, 19(6): 1528-1538.

[12] 薛俊鹏,等. 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):145-151.

Xue J P,et al. Camera calibration with single image based on two orthogonal one-dimensional objects［J］. Acta Optica Sinica ,2012,32(1):145-151.

[13] 贺科学,等.基于两垂直相交线段的摄像机快速标定算法［J］. 仪器仪表学报,2013,34(8):1696:1702.

HE K X,et al. Fast camera calibration algorithm based on two perpendicular line segments［J］. Chinese Journal of Scientific Instrument, 2013,34(8):1696:1702.

[14] 白瑞林,等.一种实用的X 型靶标亚像素角点提取方法［J］.光学技术, 2010,36(4):560-565.

BAI R L, et al. A practical method for detection of sub-pixel corners for X-target［J］. Optical Technique, 2010,36(4):560-565.

[15] BOUGUET J Y. Camera calibration toolbox for matlab［EB/OL］. MRL-Intel Corp,(2010-7-9). http:∥www. vision. caltech.edu/bouguetj/calib_doc/.

贺科学, 李树涛, 胡建文. 基于两平行线及其线上三点的摄像机标定方法[J]. 光学技术, 2016, 42(6): 545. HE Kexue, LI Shutao, HU Jianwen. Camera calibration based on two parallel lines and three collinear points[J]. Optical Technique, 2016, 42(6): 545.

基于两平行线及其线上三点的摄像机标定方法

关于本站 Cookie 的使用提示

全站搜索

基于两平行线及其线上三点的摄像机标定方法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索