A technique of camera calibration using single planar calibration image

Abstract

The paper presents a novel camera calibration method of intrinsic and extrinsic parameters based on single planar calibration image. A camera model including backward radial nonlinear model and indirect extrinsic parameters model is proposed at first, and then constraint equations are constructed by all distances between control points, followed by a nonlinear iterative optimization step to obtain all camera parameters. The experimental results and theoretical analysis show that, if there is obviously radial nonlinearity in calibration image, the intrinsic and extrinsic parameters can be estimated from single planar calibration image, and if there are multiple-view planar calibration images, the results are slightly better than classical Zhang's method.